GPQA Diamond

sakana-fugu

Sakana AI

closed

Source carrier

95.5%

sakana-fugu-ultra

Sakana AI

closed

Source carrier

95.5%

grok-4.6

xAI

closed

Score input

94.9%

gpt-5.4

OpenAI

closed

Public reference

94.6%

gpt-5.6-sol

OpenAI

closed

Source carrier

94.6%

claude-opus-4-7

Anthropic

closed

Source carrier

94.2%

gemini-3.1-pro-preview

Google

closed

Public reference

94.1%

gemini-3.6-flash

Google

closed

Score input

94.1%

Claude Mythos 5

claude-mythos-5

Anthropic

closed

Source carrier

94.1%

#10

gpt-5.5

OpenAI

closed

Score input

94.0%

#11

claude-opus-5

Anthropic

closed

Score input

93.9%

#12

claude-opus-4-8

Anthropic

closed

Source carrier

93.6%

#13

kimi-k3

Moonshot AI

open

Score input

93.5%

#14

grok-4.5

xAI

closed

Score input

93.4%

#15

gpt-5.6-terra

OpenAI

closed

Score input

93.3%

#16

MiniMax-M3

MiniMax

open

Public reference

92.9%

#17

DeepSeek V4 Pro 0813

DeepSeek-V4-Pro-0813

DeepSeek

open

Public reference

92.8%

#18

gemini-3.5-flash

Google

closed

Score input

92.8%

#19

Qwen3.8 Max

qwen3.8-max

Alibaba Cloud

open

Score input

92.7%

#20

claude-fable-5

Anthropic

closed

Public reference

92.6%

#21

qwen3.7-max

Alibaba Cloud

closed

Source carrier

92.4%

#22

Qwen3.8-Flash-Next

Qwen/Qwen3.8-Flash-Next

Alibaba Cloud

open

Public reference

92.3%

#23

gpt-5.6-luna

OpenAI

closed

Source carrier

92.3%

#24

Gemini 3.7 Flash

gemini-3.7-flash

Google

closed

Score input

92.1%

#25

glm-5-2

Z.AI

open

Score input

91.9%

#26

GLM-5.3

glm-5.3

Z.AI

announced-open

Score input

91.7%

#27

gpt-5.3-codex

OpenAI

closed

Public reference

91.5%

#28

DeepSeek V4 Flash Vision Exp

gpt-5.2

OpenAI

closed

Score input

91.4%

#29

deepseek-v4-flash-vision-exp

DeepSeek

—

Public reference

91.3%

#30

claude-opus-4-6

Anthropic

closed

Source carrier

91.3%

#31

GLM-5.3-Flash

glm-5.3-flash

Z.AI

open

Score input

91.2%

#32

claude-sonnet-5

Anthropic

closed

Score input

91.1%

#33

grok-4.20

xAI

closed

Public reference

91.1%

#34

kimi-k2.6

Moonshot AI

open

Public reference

91.1%

#35

Muse Spark 1.2

muse-spark-1-2

Meta

announced-open

Score input

90.4%

#36

hy3

Tencent

open

Score input

90.4%

#37

qwen3-6-plus

Alibaba Cloud

closed

Source carrier

90.4%

#38

qwen3.7-plus

Alibaba Cloud

closed

Source carrier

90.3%

#39

deepseek-v4-pro

DeepSeek

open

Source carrier

90.1%

#40

grok-4.3

xAI

closed

Source carrier

90.1%

#41

claude-sonnet-4-6

Anthropic

closed

Source carrier

89.9%

#42

gpt-5.2-codex

OpenAI

closed

Source carrier

89.9%

#43

interfaze-beta

Interfaze

closed

Source carrier

89.9%

#44

muse-spark-1-1

Meta

closed

Public reference

89.8%

#45

kimi-k2.7-code

Moonshot AI

open

Source carrier

89.6%

#46

Inkling-Small

thinkingmachines/inkling-small

Thinking Machines Lab

open

Source carrier

89.5%

#47

muse-spark

Meta

closed

Source carrier

89.5%

#48

deepseek-v4-flash

DeepSeek

open

Public reference

89.4%

#49

Qwen/Qwen3.5-397B-A17B

Alibaba Cloud

open

Source carrier

89.3%

#50

Qwen3.8-27B

Qwen/Qwen3.8-27B

Alibaba Cloud

open

Score input

89.2%

#51

LongCat-2.0

meituan-longcat/LongCat-2.0

Meituan

open

Score input

88.9%

#52

qwen3-6-max-preview

Alibaba Cloud

closed

Source carrier

88.8%

#53

Qwen3.6 Max

qwen3.6-max

Alibaba Cloud

closed

Public reference

88.8%

#54

inkling

Thinking Machines Lab

open

Score input

88.3%

#55

kimi-k2.5

Moonshot AI

open

Source carrier

87.9%

#56

grok-4

xAI

closed

Source carrier

87.7%

#57

gpt-5.1

OpenAI

closed

Score input

87.6%

#58

gpt-5.4-mini

OpenAI

closed

Source carrier

87.5%

#59

hy3-preview

Tencent

open

Source carrier

87.2%

#60

claude-opus-4-5

Anthropic

closed

Source carrier

87%

#61

mimo-v2-pro

Xiaomi

closed

Source carrier

87%

#62

minimax-m2-7

MiniMax

open

Source carrier

87%

#63

nemotron-3-ultra-550b

NVIDIA

open

Source carrier

87%

#64

claude-opus-4-5-thinking

Anthropic

closed

Source carrier

86.6%

#65

mimo-v2-5-pro

Xiaomi

open

Source carrier

86.6%

#66

Solar Open 2 250B

upstage/Solar-Open2-250B

Upstage

open

Score input

86.3%

#67

glm-5-1

Z.AI

open

Source carrier

86.2%

#68

qwen3-max

Alibaba Cloud

closed

Public reference

86.1%

#69

glm-5

Z.AI

open

Source carrier

86%

#70

gpt-5-1-codex

OpenAI

closed

Source carrier

86%

#71

gpt-5-1-codex-max

OpenAI

closed

Source carrier

86%

#72

glm-4.7

Z.AI

open

Source carrier

85.9%

#73

qwen3.5-27b

Alibaba Cloud

open

Source carrier

85.8%

#74

gemma-4-31b

Google

open

Source carrier

85.7%

#75

Qwen/Qwen3.5-122B-A10B

Alibaba Cloud

open

Source carrier

85.7%

#76

Qwen3.5 122B

qwen3.5-122b-a10b

Alibaba Cloud

open

Score input

85.7%

#77

gpt-5

OpenAI

closed

Source carrier

85.4%

#78

grok-4-1-fast-reasoning

xAI

closed

Source carrier

85.3%

#79

MiMo-V2.5

mimo-v2.5

Xiaomi

open

Public reference

84.9%

#80

MiniMax M2.5

minimax-m2.5

MiniMax

open

Public reference

84.8%

#81

Grok 4 Fast

grok-4-fast

xAI

closed

Public reference

84.7%

#82

glm-5-turbo

Z.AI

closed

Source carrier

84.7%

#83

grok-4-fast-reasoning

xAI

closed

Source carrier

84.7%

#84

OpenAI

closed

Public reference

84.5%

#85

Qwen3.5 35B

qwen3.5-35b-a3b

Alibaba Cloud

open

Score input

84.5%

#86

qwen3-5-35b-a3b

Alibaba Cloud

open

Source carrier

84.5%

#87

gemini-2.5-pro

Google

closed

Public reference

84.4%

#88

qwen3.6-27b

Alibaba Cloud

open

Public reference

84.2%

#89

mai-thinking-1

Microsoft

closed

Source carrier

84.2%

#90

Qwen/Qwen3.6-35B-A3B

Alibaba Cloud

open

Score input

84.1%

#91

deepseek-v3.2

DeepSeek

open

Public reference

84.0%

#92

Kimi K2 Thinking

kimi-k2-thinking

Moonshot AI

open

Public reference

83.8%

#93

gemini-3.5-flash-lite

Google

closed

Public reference

83.8%

#94

GPT-5 Codex

gpt-5-codex

OpenAI

closed

Public reference

83.7%

#95

Muse Glimmer 30B

meta-models/Muse-Glimmer-30B

Meta

open

Score input

83.5%

#96

claude-sonnet-4-5

Anthropic

closed

Score input

83.4%

#97

MiniMax M2.1

minimax-m2.1

MiniMax

open

Public reference

83.0%

#98

gpt-5.4-nano

OpenAI

closed

Source carrier

82.8%

#99

mimo-v2-omni

Xiaomi

closed

Source carrier

82.8%

#100

OpenAI

closed

Public reference

82.7%

#101

Qwen3.5 Omni Plus

qwen3.5-omni-plus

Alibaba Cloud

closed

Public reference

82.6%

#102

gemini-3.1-flash-lite

Google

closed

Source carrier

82.2%

#103

deepseek-r1

DeepSeek

open

Source carrier

81.3%

#104

gemini-3-flash

Google

closed

Source carrier

81.2%

#105

claude-4-1-opus-thinking

Anthropic

closed

Source carrier

80.9%

#106

claude-opus-4-1

Anthropic

closed

Score input

80.9%

#107

glm-5v-turbo

Z.AI

closed

Source carrier

80.9%

#108

step-3-7-flash

StepFun

open

Source carrier

80.9%

#109

gpt-5-mini

OpenAI

closed

Public reference

80.3%

#110

Nemotron 3 Super

nemotron-3-super-120b

NVIDIA

open

Score input

80%

#111

claude-opus-4

Anthropic

closed

Score input

79.6%

#112

gemini-2.5-flash

Google

closed

Public reference

79.3%

#113

google/gemma-4-26b-a4b-it

Google

open

Source carrier

79.2%

#114

DeepSeek V3.1 Terminus

deepseek-v3.1-terminus

DeepSeek

open

Public reference

79.2%

#115

Gemma 4 26B

gemma-4-26b-a4b

Google

open

Score input

79.2%

#116

grok-3-mini

xAI

closed

Score input

79.1%

#117

google/gemma-4-12B

Google

open

Source carrier

78.8%

#118

Nova 2 Pro

nova-2-pro-preview

Amazon

closed

Score input

78.5%

#119

OpenAI

closed

Public reference

78.4%

#120

k-exaone

LG AI Research

closed

Source carrier

78.3%

#121

gpt-oss-120b

OpenAI

open

Source carrier

78.2%

#122

glm-4.6

Z.AI

open

Public reference

78.0%

#123

deepseek-v3-1

DeepSeek

open

Source carrier

77.9%

#124

claude-sonnet-4

Anthropic

closed

Score input

77.7%

#125

MiniMax M2

minimax-m2

MiniMax

open

Public reference

77.7%

#126

Claude Sonnet 3.7

claude-3-7-sonnet

Anthropic

closed

Score input

77.2%

#127

mistral-small-2603

Mistral AI

open

Source carrier

76.9%

#128

Kimi K2 0905

kimi-k2-0905

Moonshot AI

open

Public reference

76.7%

#129

kimi-k2

Moonshot AI

closed

Source carrier

76.6%

#130

trinity-large-thinking

Arcee AI

open

Source carrier

76.3%

#131

command-a-plus

Cohere

open

Source carrier

76.1%

#132

NVIDIA Nemotron 3.5 Lightning 30B-A3B

nemotron-3-nano-30b

NVIDIA

open

Source carrier

75.7%

#133

nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16

NVIDIA

open

Score input

75.4%

#134

Ling 2.6 1T

ling-2.6-1t

InclusionAI

open

Public reference

75.2%

#135

Mistral Medium 3.5

mistral-medium-3-5

Mistral AI

open

Public reference

74.8%

#136

mistralai/Mistral-Medium-3.5-128B

Mistral AI

open

Score input

74.8%

#137

OpenAI

closed

Source carrier

74.8%

#138

OpenAI

closed

Public reference

74.7%

#139

sarvam-105b

Sarvam

open

Source carrier

73.8%

#140

glm-4-5-air

Z.AI

closed

Source carrier

73.3%

#141

nemotron-ultra-253b

NVIDIA

open

Source carrier

72.8%

#142

grok-code-fast-1

xAI

closed

Public reference

72.7%

#143

nemotron-3-nano-omni-30b-a3b

NVIDIA

open

Source carrier

72.2%

#144

zaya1-8b

Zyphra

open

Source carrier

71%

#145

gpt-5-nano

OpenAI

closed

Score input

69.4%

#146

gpt-oss-20b

OpenAI

open

Source carrier

68.8%

#147

claude-4-sonnet

Anthropic

closed

Source carrier

68.3%

#148

mistral-large-2512

Mistral AI

open

Source carrier

68%

#149

llama-4-maverick

Meta

open

Source carrier

67.1%

#150

gpt-4-1

OpenAI

closed

Source carrier

66.6%

#151

gpt-4-1-mini

OpenAI

closed

Source carrier

66.4%

#152

claude-haiku-4-5-20251001

Anthropic

closed

Score input

64.7%

#153

grok-4-1-fast

xAI

closed

Source carrier

63.7%

#154

sarvam-30b

Sarvam

open

Source carrier

63.3%

#155

trinity-large-preview

Arcee AI

open

Source carrier

63.3%

#156

exaone-4-0-32b

LG AI Research

open

Source carrier

62.8%

#157

deepseek-r1-distill-qwen-32b

DeepSeek

open

Source carrier

61.5%

#158

ling-2-6-flash

InclusionAI

open

Source carrier

59.3%

#159

gemini-1-5-pro

Google

closed

Source carrier

58.9%

#160

llama-4-scout

Meta

open

Source carrier

58.7%

#161

mistral-medium-3

Mistral AI

closed

Source carrier

57.8%

#162

gemma-4-e4b

Google

open

Source carrier

57.6%

#163

mellum2-12b-a2-5b-thinking

JetBrains

open

Source carrier

57.6%

#164

phi-4

Microsoft

open

Source carrier

57.5%

#165

zaya1-74b-preview

Zyphra

open

Source carrier

57.3%

#166

solar-pro-2

Upstage

closed

Source carrier

56.1%

#167

deepseek-v3

DeepSeek

open

Source carrier

55.7%

#168

gpt-4o

OpenAI

closed

Source carrier

54.3%

#169

llama-3-1-405b

Meta

open

Source carrier

51.5%

#170

lfm2-5-8b-a1b

LiquidAI

open

Source carrier

51.3%

#171

gpt-4-1-nano

OpenAI

closed

Score input

51.2%

#172

nova-pro

Amazon

closed

Source carrier

49.9%

#173

claude-3-opus

Anthropic

closed

Source carrier

48.9%

#174

mistral-large-2

Mistral AI

closed

Source carrier

48.6%

#175

soofi-s-30b-a3b

Soofi Project

open

Source carrier

43.4%

#176

gemma-3-27b

Google

open

Score input

42.8%

#177

gpt-4o-mini

OpenAI

closed

Source carrier

42.6%

#178

exaone-4-0-1-2b

LG AI Research

open

Source carrier

42.4%

#179

qwen2-5-coder-32b-instruct

Alibaba Cloud

open

Source carrier

41.7%

#180

mellum2-12b-a2-5b-instruct

JetBrains

open

Source carrier

40.9%

#181

gemma-4-e2b

Google

open

Source carrier

40.5%

#182

claude-3-haiku

Anthropic

closed

Source carrier

37.4%

#183

granite-4-0-h-350m

IBM

open

Source carrier

29%

#184

lfm2-5-vl-1-6b-extract

LiquidAI

open

Source carrier

28.9%

#185

granite-4-0-1b

IBM

open

Source carrier

28.1%

#186

gemini-1-0-pro

Google

closed

Source carrier

27.7%

#187

granite-4-0-h-1b

IBM

open

Source carrier

26.3%

#188

minicpm5-1b

OpenBMB

open

Source carrier

26.3%

#189

lfm2-5-230m

LiquidAI

open

Source carrier

25.4%

#190

granite-4-0-350m

IBM

open

Source carrier

23.7%

Every distinct dated source-published row for GPQA Diamond, including exact model configuration labels
Exact published model / configuration	Canonical product	Score	Version / system	Evidence	Source / observed
Qwen3.8-Flash-Next (Artificial Analysis independent run) qwen-3-8-flash-next-aa-unspecified	Qwen3.8-Flash-Next	92.3%	diamondArtificial Analysis GPQA Diamond evaluation	Public reference · not scored independently-verified	Qwen3.8-Flash-Next individual evaluations ↗Observed 2026-08-29 · checked 2026-08-29
Claude Opus 4.7 (Adaptive Reasoning, Max Effort) (Artificial Analysis independent run) claude-opus-4-7-max	Claude Opus 4.7	91.4%	diamondArtificial Analysis GPQA Diamond evaluation	Score input independently-verified	Claude Opus 4.7 (Adaptive Reasoning, Max Effort) individual evaluations ↗Observed 2026-08-29 · checked 2026-08-29
Claude Opus 4.6 (Adaptive Reasoning, Max Effort) (Artificial Analysis independent run) claude-opus-4-6-max	Claude Opus 4.6	89.6%	diamondArtificial Analysis GPQA Diamond evaluation	Public reference · not scored independently-verified	Claude Opus 4.6 (Adaptive Reasoning, Max Effort) individual evaluations ↗Observed 2026-08-29 · checked 2026-08-29
GLM-5.2 (max) (Artificial Analysis independent run) glm-5-2-max	GLM-5.2	89.5%	diamondArtificial Analysis GPQA Diamond evaluation	Score input independently-verified	GLM-5.2 (max) individual evaluations ↗Observed 2026-08-29 · checked 2026-08-29
Qwen3.5 397B A17B (Reasoning) (Artificial Analysis completed independent run) qwen3-5-397b-thinking	Qwen3.5 397B A17B	89.3%	diamondArtificial Analysis GPQA Diamond evaluation	Score input independently-verified	Qwen3.5 397B A17B (Reasoning) current model and individual-evaluation page ↗Observed 2026-08-29 · checked 2026-08-29
Kimi K2.5 (Reasoning) (Artificial Analysis completed independent run) kimi-k2-5-thinking	Kimi K2.5	87.9%	diamondArtificial Analysis GPQA Diamond evaluation	Score input independently-verified	Kimi K2.5 (Reasoning) current model and individual-evaluation page ↗Observed 2026-08-29 · checked 2026-08-29
Qwen3.5 122B A10B (Reasoning) (Artificial Analysis completed independent run) qwen3-5-122b-a10b-aa-reasoning-default	Qwen3.5-122B-A10B	85.7%	diamondArtificial Analysis GPQA Diamond evaluation	Score input independently-verified	Qwen3.5 122B A10B (Reasoning) current model and individual-evaluation page ↗Observed 2026-08-29 · checked 2026-08-29
Qwen3.6 35B A3B (Reasoning) (Artificial Analysis completed independent run) qwen3-6-35b-a3b-aa-reasoning-default	Qwen3.6-35B-A3B	84.1%	diamondArtificial Analysis GPQA Diamond evaluation	Score input independently-verified	Qwen3.6 35B A3B (Reasoning) current model and individual-evaluation page ↗Observed 2026-08-29 · checked 2026-08-29
Gemma 4 26B A4B (Reasoning) (Artificial Analysis completed independent run) gemma-4-26b-a4b-aa-reasoning-default	Gemma 4 26B A4B	79.2%	diamondArtificial Analysis GPQA Diamond evaluation	Score input independently-verified	Gemma 4 26B A4B (Reasoning) current model and individual-evaluation page ↗Observed 2026-08-29 · checked 2026-08-29
LongCat 2.0 (Artificial Analysis independent run) longcat-2-0-default	LongCat-2.0	78.0%	diamondArtificial Analysis GPQA Diamond evaluation	Score input independently-verified	LongCat 2.0 individual evaluations ↗Observed 2026-08-29 · checked 2026-08-29
DeepSeek V3.1 (Reasoning) (Artificial Analysis independent run) deepseek-v3-1-reasoning-default	DeepSeek V3.1	77.9%	diamondArtificial Analysis GPQA Diamond evaluation	Public reference · not scored independently-verified	DeepSeek V3.1 (Reasoning) individual evaluations ↗Observed 2026-08-29 · checked 2026-08-29
Mistral Small 4 (Reasoning) (Artificial Analysis completed independent run) mistral-small-4-reasoning	Mistral Small 4	76.9%	diamondArtificial Analysis GPQA Diamond evaluation	Score input independently-verified	Mistral Small 4 (Reasoning) current model and individual-evaluation page ↗Observed 2026-08-29 · checked 2026-08-29
NVIDIA Nemotron 3 Nano 30B A3B (Reasoning) (Artificial Analysis completed independent run) nemotron-3-nano-30b-aa-reasoning-default	Nemotron 3 Nano 30B	75.7%	diamondArtificial Analysis GPQA Diamond evaluation	Score input independently-verified	NVIDIA Nemotron 3 Nano 30B A3B (Reasoning) current model and individual-evaluation page ↗Observed 2026-08-29 · checked 2026-08-29
Trinity Large Thinking (Artificial Analysis completed independent run) trinity-large-thinking-aa-reasoning-default	Trinity-Large-Thinking	75.2%	diamondArtificial Analysis GPQA Diamond evaluation	Score input independently-verified	Trinity Large Thinking current model and individual-evaluation page ↗Observed 2026-08-29 · checked 2026-08-29
Mistral Medium 3.5 (Artificial Analysis completed independent run) mistral-medium-3-5-128b-aa-reasoning-default	Mistral Medium 3.5 128B	74.8%	diamondArtificial Analysis GPQA Diamond evaluation	Score input independently-verified	Mistral Medium 3.5 current model and individual-evaluation page ↗Observed 2026-08-29 · checked 2026-08-29
DeepSeek V3.1 (Non-reasoning) (Artificial Analysis independent run) deepseek-v3-1-non-reasoning	DeepSeek V3.1	73.5%	diamondArtificial Analysis GPQA Diamond evaluation	Score input independently-verified	DeepSeek V3.1 (Non-reasoning) individual evaluations ↗Observed 2026-08-29 · checked 2026-08-29
Llama 4 Maverick (Artificial Analysis completed independent run) llama-4-maverick-aa-non-reasoning-default	Llama 4 Maverick	67.1%	diamondArtificial Analysis GPQA Diamond evaluation	Score input independently-verified	Llama 4 Maverick current model and individual-evaluation page ↗Observed 2026-08-29 · checked 2026-08-29
GPT-4.1 mini (Artificial Analysis completed independent run) gpt-4-1-mini-epoch-gpt-4-1-mini-2025-04-14	GPT-4.1 mini	66.4%	diamondArtificial Analysis GPQA Diamond evaluation	Score input independently-verified	GPT-4.1 mini current model and individual-evaluation page ↗Observed 2026-08-29 · checked 2026-08-29
Llama 4 Scout (Artificial Analysis completed independent run) llama-4-scout-aa-non-reasoning-default	Llama 4 Scout	58.7%	diamondArtificial Analysis GPQA Diamond evaluation	Score input independently-verified	Llama 4 Scout current model and individual-evaluation page ↗Observed 2026-08-29 · checked 2026-08-29
GPT-4.1 nano (Artificial Analysis completed independent run) gpt-4-1-nano-epoch-gpt-4-1-nano-2025-04-14	GPT-4.1 nano	51.2%	diamondArtificial Analysis GPQA Diamond evaluation	Score input independently-verified	GPT-4.1 nano current model and individual-evaluation page ↗Observed 2026-08-29 · checked 2026-08-29
Gemma 3 27B Instruct (Artificial Analysis completed independent run) gemma-3-27b-aa-non-reasoning-default	Gemma 3 27B	42.8%	diamondArtificial Analysis GPQA Diamond evaluation	Score input independently-verified	Gemma 3 27B Instruct current model and individual-evaluation page ↗Observed 2026-08-29 · checked 2026-08-29
Grok 4.6 (high; Artificial Analysis independent run) grok-4-6-high	Grok 4.6	94.9%	diamondArtificial Analysis GPQA Diamond evaluation	Score input independently-verified	Grok 4.6 individual evaluations ↗Observed 2026-08-27 · checked 2026-08-27
GPT-5.6 Sol (max; Artificial Analysis independent run) gpt-5-6-sol-max	GPT-5.6 Sol	94.1%	diamondArtificial Analysis GPQA Diamond evaluation	Score input independently-verified	GPT-5.6 Sol individual evaluations ↗Observed 2026-08-27 · checked 2026-08-27
GPT-5.5 (xhigh; Artificial Analysis independent run) gpt-5-5-xhigh	GPT-5.5	93.5%	diamondArtificial Analysis GPQA Diamond evaluation	Score input independently-verified	GPT-5.5 individual evaluations ↗Observed 2026-08-27 · checked 2026-08-27
Kimi K3 (max; Artificial Analysis independent run) kimi-k3-max	Kimi K3	93.5%	diamondArtificial Analysis GPQA Diamond evaluation	Score input independently-verified	Kimi K3 individual evaluations ↗Observed 2026-08-27 · checked 2026-08-27
Claude Opus 5 (max; Artificial Analysis independent run) claude-opus-5-max	Claude Opus 5	93.2%	diamondArtificial Analysis GPQA Diamond evaluation	Score input independently-verified	Claude Opus 5 individual evaluations ↗Observed 2026-08-27 · checked 2026-08-27
Grok 4.5 (high; Artificial Analysis independent run) grok-4-5-aa-2-high	Grok 4.5	93.1%	diamondArtificial Analysis GPQA Diamond evaluation	Score input independently-verified	Grok 4.5 individual evaluations ↗Observed 2026-08-27 · checked 2026-08-27
DeepSeek V4 Pro 0813 (max; Artificial Analysis independent run) deepseek-v4-pro-0813-max	DeepSeek V4 Pro 0813	92.8%	diamondArtificial Analysis GPQA Diamond evaluation	Public reference · not scored independently-verified	DeepSeek V4 Pro 0813 individual evaluations ↗Observed 2026-08-27 · checked 2026-08-27
Gemini 3.6 Flash (high; Artificial Analysis independent run) gemini-3-6-flash-high	Gemini 3.6 Flash	92.8%	diamondArtificial Analysis GPQA Diamond evaluation	Score input independently-verified	Gemini 3.6 Flash individual evaluations ↗Observed 2026-08-27 · checked 2026-08-27
Claude Fable 5 (max; Artificial Analysis independent run) claude-fable-5-max	Claude Fable 5	92.6%	diamondArtificial Analysis GPQA Diamond evaluation	Score input independently-verified	Claude Fable 5 individual evaluations ↗Observed 2026-08-27 · checked 2026-08-27
GPT-5.6 Terra (max; Artificial Analysis independent run) gpt-5-6-terra-max	GPT-5.6 Terra	92.5%	diamondArtificial Analysis GPQA Diamond evaluation	Score input independently-verified	GPT-5.6 Terra individual evaluations ↗Observed 2026-08-27 · checked 2026-08-27
Gemini 3.7 Flash (medium; Artificial Analysis independent run) gemini-3-7-flash-medium	Gemini 3.7 Flash	92.1%	diamondArtificial Analysis GPQA Diamond evaluation	Score input independently-verified	Gemini 3.7 Flash individual evaluations ↗Observed 2026-08-27 · checked 2026-08-27
Claude Opus 4.8 (max; Artificial Analysis independent run) claude-opus-4-8-max	Claude Opus 4.8	92.0%	diamondArtificial Analysis GPQA Diamond evaluation	Score input independently-verified	Claude Opus 4.8 individual evaluations ↗Observed 2026-08-27 · checked 2026-08-27
GPT-5.4 (xhigh; Artificial Analysis independent run) gpt-5-4-xhigh	GPT-5.4	92.0%	diamondArtificial Analysis GPQA Diamond evaluation	Score input independently-verified	GPT-5.4 individual evaluations ↗Observed 2026-08-27 · checked 2026-08-27
GLM-5.3 (max; Artificial Analysis independent run) glm-5-3-max	GLM-5.3	91.7%	diamondArtificial Analysis GPQA Diamond evaluation	Score input independently-verified	GLM-5.3 individual evaluations ↗Observed 2026-08-27 · checked 2026-08-27
DeepSeek V4 Flash Vision Exp (max; Artificial Analysis independent run) deepseek-v4-flash-vision-exp-max-harness	DeepSeek V4 Flash Vision Exp	91.3%	diamondArtificial Analysis GPQA Diamond evaluation	Public reference · not scored independently-verified	DeepSeek V4 Flash Vision Exp individual evaluations ↗Observed 2026-08-27 · checked 2026-08-27
GLM-5.3-Flash (max; Artificial Analysis independent run) glm-5-3-flash-max	GLM-5.3-Flash	91.2%	diamondArtificial Analysis GPQA Diamond evaluation	Score input independently-verified	GLM-5.3-Flash individual evaluations ↗Observed 2026-08-27 · checked 2026-08-27
Claude Sonnet 5 (max; Artificial Analysis independent run) claude-sonnet-5-max	Claude Sonnet 5	91.1%	diamondArtificial Analysis GPQA Diamond evaluation	Score input independently-verified	Claude Sonnet 5 individual evaluations ↗Observed 2026-08-27 · checked 2026-08-27
GPT-5.6 Luna (max; Artificial Analysis independent run) gpt-5-6-luna-max	GPT-5.6 Luna	91.1%	diamondArtificial Analysis GPQA Diamond evaluation	Score input independently-verified	GPT-5.6 Luna individual evaluations ↗Observed 2026-08-27 · checked 2026-08-27
Muse Spark 1.2 (xhigh; Artificial Analysis independent run) muse-spark-1-2-xhigh	Muse Spark 1.2	90.4%	diamondArtificial Analysis GPQA Diamond evaluation	Score input independently-verified	Muse Spark 1.2 individual evaluations ↗Observed 2026-08-27 · checked 2026-08-27
Muse Spark 1.1 (xhigh; Artificial Analysis independent run) muse-spark-1-1-xhigh	Muse Spark 1.1	89.8%	diamondArtificial Analysis GPQA Diamond evaluation	Score input independently-verified	Muse Spark 1.1 individual evaluations ↗Observed 2026-08-27 · checked 2026-08-27
Qwen3.8-Flash-Next (xhigh default thinking configuration) qwen-3-8-flash-next-xhigh	Qwen3.8-Flash-Next	91.7%	Diamondsystem:qwen3-8-flash-next:gpqa	Score input source-checked	Qwen3.8-Flash-Next current official launch page ↗Observed 2026-08-26 · checked 2026-08-29
Claude Opus 4.6 (Max) as published in Qwen's Qwen3.8-Flash-Next comparison table claude-opus-4-6-max	Claude Opus 4.6	91.3%	Diamondsystem:qwen3-8-flash-next-comparison:cell:language:gpqa:4	Relative comparison · not standard direct provider-reported	Qwen3.8-Flash-Next launch and official provider evaluations ↗Observed 2026-08-26 · checked 2026-08-26
DeepSeek-V4-Flash-0731 (provider-published configuration) as published in Qwen's Qwen3.8-Flash-Next comparison table deepseek-v4-flash-0731-deepseek-0813-release-unspecified	DeepSeek V4 Flash 0731	90.8%	Diamondsystem:qwen3-8-flash-next-comparison:cell:language:gpqa:3	Relative comparison · not standard direct provider-reported	Qwen3.8-Flash-Next launch and official provider evaluations ↗Observed 2026-08-26 · checked 2026-08-26
Qwen3.7-Plus (provider-published configuration) as published in Qwen's Qwen3.8-Flash-Next comparison table qwen-3-7-plus-unspecified	Qwen3.7-Plus	90.3%	Diamondsystem:qwen3-8-flash-next-comparison:cell:language:gpqa:2	Relative comparison · not standard direct provider-reported	Qwen3.8-Flash-Next launch and official provider evaluations ↗Observed 2026-08-26 · checked 2026-08-26
Qwen3.8-27B (xhigh default thinking configuration) as published in Qwen's Qwen3.8-Flash-Next comparison table qwen-3-8-27b-xhigh	Qwen3.8-27B	89.2%	Diamondsystem:qwen3-8-flash-next-comparison:cell:language:gpqa:1	Relative comparison · not standard direct provider-reported	Qwen3.8-Flash-Next launch and official provider evaluations ↗Observed 2026-08-26 · checked 2026-08-26
GLM-5.3 (max) glm-5-3-max	GLM-5.3	90.9%	1.0.11Epoch AI benchmark runner	Score input independently-verified	Epoch AI permanent refresh source ↗Observed 2026-08-24 · checked 2026-09-01
Claude Opus 4.6 Max as published by Qwen claude-opus-4-6-max	Claude Opus 4.6	91.3%	DiamondQwen3.8-27B official text table	Relative comparison · not standard direct provider-reported	Qwen3.8-27B official model card ↗Observed 2026-08-15 · checked 2026-08-15
Qwen3.7-Plus as published by Qwen qwen-3-7-plus-unspecified	Qwen3.7-Plus	90.3%	DiamondQwen3.8-27B official text table	Relative comparison · not standard direct provider-reported	Qwen3.8-27B official model card ↗Observed 2026-08-15 · checked 2026-08-15
Qwen3.8-27B (xhigh) qwen-3-8-27b-xhigh	Qwen3.8-27B	89.2%	DiamondQwen3.8-27B official text table	Score input provider-reported	Qwen3.8-27B official model card ↗Observed 2026-08-15 · checked 2026-08-15
DeepSeek V4 Flash max as published by Upstage deepseek-v4-flash-max	DeepSeek V4 Flash	88.9%	DiamondSolar Open 2 card English table	Relative comparison · not standard direct provider-reported	Solar Open 2 250B model card ↗Observed 2026-08-15 · checked 2026-08-15
Qwen3.6-27B as published by Qwen qwen3-6-27b-default	Qwen3.6 27B	87.8%	DiamondQwen3.8-27B official text table	Relative comparison · not standard direct provider-reported	Qwen3.8-27B official model card ↗Observed 2026-08-15 · checked 2026-08-15
Solar Open 2 250B (high) solar-open2-250b-high	Solar Open 2 250B	86.3%	DiamondSolar Open 2 card English table	Score input provider-reported	Solar Open 2 250B model card ↗Observed 2026-08-15 · checked 2026-08-15
MiMo-V2.5 as published by Upstage mimo-v2-5-solar-open2-unspecified	MiMo-V2.5	83%	DiamondSolar Open 2 card English table	Relative comparison · not standard direct provider-reported	Solar Open 2 250B model card ↗Observed 2026-08-15 · checked 2026-08-15
Mistral Medium 3.5 as published by Upstage mistral-medium-3-5-high	Mistral Medium 3.5	77.5%	DiamondSolar Open 2 card English table	Relative comparison · not standard direct provider-reported	Solar Open 2 250B model card ↗Observed 2026-08-15 · checked 2026-08-15
Command A+ as published by Upstage command-a-plus-solar-open2-unspecified	Command A+	75.6%	DiamondSolar Open 2 card English table	Relative comparison · not standard direct provider-reported	Solar Open 2 250B model card ↗Observed 2026-08-15 · checked 2026-08-15
Solar Open 100B as published by Upstage solar-open-100b-reasoning-high	Solar Open 100B (Reasoning)	66.2%	DiamondSolar Open 2 card English table	Relative comparison · not standard direct provider-reported	Solar Open 2 250B model card ↗Observed 2026-08-15 · checked 2026-08-15
Gemini 3.7 Flash (high) gemini-3-7-flash-high	Gemini 3.7 Flash	94.8%	1.0.11Epoch AI benchmark runner	Public reference · not scored source-checked	Epoch AI permanent refresh source ↗Observed 2026-08-14 · checked 2026-08-17
Grok 4.6 (xhigh) grok-4-6-xhigh	Grok 4.6	93.2%	1.0.11Epoch AI benchmark runner	Public reference · not scored source-checked	Epoch AI permanent refresh source ↗Observed 2026-08-14 · checked 2026-08-17
Inkling Small (xhigh) inkling-small-epoch-inkling-small-xhigh	Inkling-Small	88.5%	1.0.11Epoch AI benchmark runner	Public reference · not scored source-checked	Epoch AI permanent refresh source ↗Observed 2026-08-14 · checked 2026-08-18
Grok 4.6 (high) grok-4-6-high	Grok 4.6	94.0%	1.0.11Epoch AI benchmark runner	Score input source-checked	Epoch AI permanent refresh source ↗Observed 2026-08-12 · checked 2026-08-17
NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16; NVIDIA release evaluation; temperature=1.0; top_p=0.95 nemotron-3-5-lightning-30b-a3b-default	NVIDIA Nemotron 3.5 Lightning 30B-A3B	75.4%	CurrentNVIDIA NeMo Gym / NeMo Evaluator SDK consistent release harness	Score input provider-reported	NVIDIA Nemotron 3.5 Lightning 30B-A3B BF16 model card ↗Observed 2026-08-11 · checked 2026-08-12
MiniMax-M3 minimax-m3-epoch-minimax-m3	MiniMax M3	90.9%	1.0.11Epoch AI benchmark runner	Public reference · not scored source-checked	Epoch AI permanent refresh source ↗Observed 2026-08-10 · checked 2026-08-17
GLM-5.1 glm-5-1-epoch-glm-5-1	GLM-5.1	89.9%	1.0.11Epoch AI benchmark runner	Public reference · not scored source-checked	Epoch AI permanent refresh source ↗Observed 2026-08-10 · checked 2026-08-17
Muse Glimmer-30B; high reasoning; temperature=1.0; top_p=0.95; top_k=64 muse-glimmer-30b-high	Muse Glimmer 30B	83.5%	198-question Diamond setArtificial Analysis GPQA Diamond documented evaluation	Public reference · not scored provider-reported	Muse Glimmer Evaluation Methodology ↗Observed 2026-08-10 · checked 2026-08-10
GLM-5.2 (no thinking) glm-5-2-epoch-glm-5-2-none	GLM-5.2	71.2%	1.0.11Epoch AI benchmark runner	Public reference · not scored source-checked	Epoch AI permanent refresh source ↗Observed 2026-08-10 · checked 2026-08-17
Kimi K3 (high) kimi-k3-epoch-kimi-k3-high	Kimi K3	91.9%	1.0.11Epoch AI benchmark runner	Public reference · not scored source-checked	Epoch AI permanent refresh source ↗Observed 2026-08-07 · checked 2026-08-17
Qwen3.7 Max qwen-3-7-max-epoch-qwen3-7-max	Qwen3.7-Max	90.9%	1.0.11Epoch AI benchmark runner	Public reference · not scored source-checked	Epoch AI permanent refresh source ↗Observed 2026-08-07 · checked 2026-08-17
GPT-5.6 Sol (low) gpt-5-6-sol-low	GPT-5.6 Sol	89.9%	1.0.11Epoch AI benchmark runner	Public reference · not scored source-checked	Epoch AI permanent refresh source ↗Observed 2026-08-07 · checked 2026-08-17
Qwen3.7 Plus (thinking) qwen-3-7-plus-epoch-qwen3-7-plus	Qwen3.7-Plus	87.9%	1.0.11Epoch AI benchmark runner	Public reference · not scored source-checked	Epoch AI permanent refresh source ↗Observed 2026-08-07 · checked 2026-08-17
Kimi K2.7 Code kimi-k2-7-code-epoch-kimi-k2-7-code	Kimi K2.7 Code	87.9%	1.0.11Epoch AI benchmark runner	Public reference · not scored source-checked	Epoch AI permanent refresh source ↗Observed 2026-08-07 · checked 2026-08-17
GPT-5.6 Terra (low) gpt-5-6-terra-low	GPT-5.6 Terra	87.4%	1.0.11Epoch AI benchmark runner	Public reference · not scored source-checked	Epoch AI permanent refresh source ↗Observed 2026-08-07 · checked 2026-08-17
Qwen3.6 Max Preview (thinking) qwen3-6-max-epoch-qwen3-6-max-preview	Qwen3.6 Max	87.4%	1.0.11Epoch AI benchmark runner	Public reference · not scored source-checked	Epoch AI permanent refresh source ↗Observed 2026-08-07 · checked 2026-08-17
GPT-5.4 mini (xhigh) gpt-5-4-mini-xhigh	GPT-5.4 mini	86.9%	1.0.11Epoch AI benchmark runner	Score input source-checked	Epoch AI permanent refresh source ↗Observed 2026-08-07 · checked 2026-08-17
Qwen3.5 397B-A17B (no thinking) qwen3-5-397b-epoch-qwen3-5-397b-a17b-none	Qwen3.5 397B A17B	86.4%	1.0.11Epoch AI benchmark runner	Public reference · not scored source-checked	Epoch AI permanent refresh source ↗Observed 2026-08-07 · checked 2026-08-17
Gemini 3.6 Flash (minimal) gemini-3-6-flash-epoch-gemini-3-6-flash-minimal	Gemini 3.6 Flash	85.9%	1.0.11Epoch AI benchmark runner	Public reference · not scored source-checked	Epoch AI permanent refresh source ↗Observed 2026-08-07 · checked 2026-08-17
Qwen3.6 27B (thinking) qwen3-6-27b-epoch-qwen3-6-27b	Qwen3.6 27B	85.9%	1.0.11Epoch AI benchmark runner	Public reference · not scored source-checked	Epoch AI permanent refresh source ↗Observed 2026-08-07 · checked 2026-08-17
Qwen3.5 397B-A17B (thinking) qwen3-5-397b-epoch-qwen3-5-397b-a17b	Qwen3.5 397B A17B	85.9%	1.0.11Epoch AI benchmark runner	Public reference · not scored source-checked	Epoch AI permanent refresh source ↗Observed 2026-08-07 · checked 2026-08-17
Kimi K3 (low) kimi-k3-low	Kimi K3	84.8%	1.0.11Epoch AI benchmark runner	Public reference · not scored source-checked	Epoch AI permanent refresh source ↗Observed 2026-08-07 · checked 2026-08-17
Qwen3.6 27B (no thinking) qwen3-6-27b-epoch-qwen3-6-27b-none	Qwen3.6 27B	84.8%	1.0.11Epoch AI benchmark runner	Public reference · not scored source-checked	Epoch AI permanent refresh source ↗Observed 2026-08-07 · checked 2026-08-17
GPT-5.6 Sol (none) gpt-5-6-sol-epoch-gpt-5-6-sol-none	GPT-5.6 Sol	82.8%	1.0.11Epoch AI benchmark runner	Public reference · not scored source-checked	Epoch AI permanent refresh source ↗Observed 2026-08-07 · checked 2026-08-17
Qwen3.7 Flash (thinking) qwen3-7-flash-epoch-qwen3-7-flash	Qwen3.7 Flash	82.3%	1.0.11Epoch AI benchmark runner	Public reference · not scored source-checked	Epoch AI permanent refresh source ↗Observed 2026-08-07 · checked 2026-08-18
GPT-5.6 Luna (low) gpt-5-6-luna-low	GPT-5.6 Luna	82.3%	1.0.11Epoch AI benchmark runner	Public reference · not scored source-checked	Epoch AI permanent refresh source ↗Observed 2026-08-07 · checked 2026-08-17
Qwen3.7 Plus (no thinking) qwen-3-7-plus-epoch-qwen3-7-plus-none	Qwen3.7-Plus	81.8%	1.0.11Epoch AI benchmark runner	Public reference · not scored source-checked	Epoch AI permanent refresh source ↗Observed 2026-08-07 · checked 2026-08-17
o3 (medium) o3-epoch-o3-2025-04-16-medium	o3	80.8%	1.0.11Epoch AI benchmark runner	Public reference · not scored source-checked	Epoch AI permanent refresh source ↗Observed 2026-08-07 · checked 2026-08-17
Qwen3.7 Flash (no thinking) qwen3-7-flash-epoch-qwen3-7-flash-none	Qwen3.7 Flash	80.8%	1.0.11Epoch AI benchmark runner	Public reference · not scored source-checked	Epoch AI permanent refresh source ↗Observed 2026-08-07 · checked 2026-08-18
o4-mini (medium) o4-mini-epoch-o4-mini-2025-04-16-medium	o4-mini	77.8%	1.0.11Epoch AI benchmark runner	Public reference · not scored source-checked	Epoch AI permanent refresh source ↗Observed 2026-08-07 · checked 2026-08-17
GPT-5.6 Terra (none) gpt-5-6-terra-epoch-gpt-5-6-terra-none	GPT-5.6 Terra	77.3%	1.0.11Epoch AI benchmark runner	Public reference · not scored source-checked	Epoch AI permanent refresh source ↗Observed 2026-08-07 · checked 2026-08-17
GPT-5.5 (no thinking) gpt-5-5-epoch-gpt-5-5-none	GPT-5.5	77.3%	1.0.11Epoch AI benchmark runner	Public reference · not scored source-checked	Epoch AI permanent refresh source ↗Observed 2026-08-07 · checked 2026-08-17
GPT-5.4 nano (low) gpt-5-4-nano-epoch-gpt-5-4-nano-2026-03-17-low	GPT-5.4 nano	72.2%	1.0.11Epoch AI benchmark runner	Public reference · not scored source-checked	Epoch AI permanent refresh source ↗Observed 2026-08-07 · checked 2026-08-17
GPT-5 mini (medium) gpt-5-mini-medium	GPT-5 mini	71.7%	1.0.11Epoch AI benchmark runner	Public reference · not scored source-checked	Epoch AI permanent refresh source ↗Observed 2026-08-07 · checked 2026-08-17
GPT-5.1 (no thinking) gpt-5-1-epoch-gpt-5-1-2025-11-13-none	GPT-5.1	66.7%	1.0.11Epoch AI benchmark runner	Public reference · not scored source-checked	Epoch AI permanent refresh source ↗Observed 2026-08-07 · checked 2026-08-17
GPT-5.4 mini (none) gpt-5-4-mini-epoch-gpt-5-4-mini-2026-03-17-none	GPT-5.4 mini	64.1%	1.0.11Epoch AI benchmark runner	Public reference · not scored source-checked	Epoch AI permanent refresh source ↗Observed 2026-08-07 · checked 2026-08-17
GPT-5.6 Luna (none) gpt-5-6-luna-epoch-gpt-5-6-luna-none	GPT-5.6 Luna	63.6%	1.0.11Epoch AI benchmark runner	Public reference · not scored source-checked	Epoch AI permanent refresh source ↗Observed 2026-08-07 · checked 2026-08-17
GPT-5.4 nano (no thinking) gpt-5-4-nano-epoch-gpt-5-4-nano-2026-03-17-none	GPT-5.4 nano	55.6%	1.0.11Epoch AI benchmark runner	Public reference · not scored source-checked	Epoch AI permanent refresh source ↗Observed 2026-08-07 · checked 2026-08-17
Gemini 3.1 Pro Preview (high) gemini-3-1-pro-preview-livebench-2026-06-25-high	Gemini 3.1 Pro Preview	94.4%	1.0.11Epoch AI benchmark runner	Public reference · not scored source-checked	Epoch AI permanent refresh source ↗Observed 2026-08-06 · checked 2026-08-17
Claude Opus 5 claude-opus-5-epoch-claude-opus-5	Claude Opus 5	92.9%	1.0.11Epoch AI benchmark runner	Public reference · not scored source-checked	Epoch AI permanent refresh source ↗Observed 2026-08-06 · checked 2026-08-17
DeepSeek v4 (high) deepseek-v4-pro-high	DeepSeek V4 Pro	90.9%	1.0.11Epoch AI benchmark runner	Public reference · not scored source-checked	Epoch AI permanent refresh source ↗Observed 2026-08-06 · checked 2026-08-18
Gemini 3 Flash Preview (high) gemini-3-flash-epoch-gemini-3-flash-preview-high	Gemini 3 Flash	89.4%	1.0.11Epoch AI benchmark runner	Public reference · not scored source-checked	Epoch AI permanent refresh source ↗Observed 2026-08-06 · checked 2026-08-17
Gemini 3.5 Flash (low) gemini-3-5-flash-epoch-gemini-3-5-flash-low	Gemini 3.5 Flash	88.9%	1.0.11Epoch AI benchmark runner	Public reference · not scored source-checked	Epoch AI permanent refresh source ↗Observed 2026-08-06 · checked 2026-08-17
Claude Opus 4.8 (low) claude-opus-4-8-epoch-claude-opus-4-8-low	Claude Opus 4.8	88.4%	1.0.11Epoch AI benchmark runner	Public reference · not scored source-checked	Epoch AI permanent refresh source ↗Observed 2026-08-06 · checked 2026-08-17
Claude Opus 4.6 (max) claude-opus-4-6-max	Claude Opus 4.6	88.4%	1.0.11Epoch AI benchmark runner	Public reference · not scored source-checked	Epoch AI permanent refresh source ↗Observed 2026-08-06 · checked 2026-08-17
Claude Opus 5 (low) claude-opus-5-low	Claude Opus 5	87.9%	1.0.11Epoch AI benchmark runner	Public reference · not scored source-checked	Epoch AI permanent refresh source ↗Observed 2026-08-06 · checked 2026-08-17
Gemini 3.6 Flash (low) gemini-3-6-flash-epoch-gemini-3-6-flash-low	Gemini 3.6 Flash	86.4%	1.0.11Epoch AI benchmark runner	Public reference · not scored source-checked	Epoch AI permanent refresh source ↗Observed 2026-08-06 · checked 2026-08-17
Claude Opus 4.7 (max) claude-opus-4-7-max	Claude Opus 4.7	86.4%	1.0.11Epoch AI benchmark runner	Score input source-checked	Epoch AI permanent refresh source ↗Observed 2026-08-06 · checked 2026-08-17
Claude Fable 5 (max) claude-fable-5-max	Claude Fable 5	85.9%	1.0.11Epoch AI benchmark runner	Score input source-checked	Epoch AI permanent refresh source ↗Observed 2026-08-06 · checked 2026-08-17
Claude Opus 4.8 (no thinking) claude-opus-4-8-epoch-claude-opus-4-8-none	Claude Opus 4.8	85.4%	1.0.11Epoch AI benchmark runner	Public reference · not scored source-checked	Epoch AI permanent refresh source ↗Observed 2026-08-06 · checked 2026-08-17
Gemini 3.5 Flash-Lite (high) gemini-3-5-flash-lite-livebench-2026-06-25-high	Gemini 3.5 Flash-Lite	83.3%	1.0.11Epoch AI benchmark runner	Public reference · not scored source-checked	Epoch AI permanent refresh source ↗Observed 2026-08-06 · checked 2026-08-17
Claude Fable 5 (high) claude-fable-5-high	Claude Fable 5	83.3%	1.0.11Epoch AI benchmark runner	Public reference · not scored source-checked	Epoch AI permanent refresh source ↗Observed 2026-08-06 · checked 2026-08-17
Gemini 3.1 Flash-Lite (high) gemini-3-1-flash-lite-epoch-gemini-3-1-flash-lite-high	Gemini 3.1 Flash-Lite	81.8%	1.0.11Epoch AI benchmark runner	Public reference · not scored source-checked	Epoch AI permanent refresh source ↗Observed 2026-08-06 · checked 2026-08-17
Claude Sonnet 5 (max) claude-sonnet-5-max	Claude Sonnet 5	80.3%	1.0.11Epoch AI benchmark runner	Score input source-checked	Epoch AI permanent refresh source ↗Observed 2026-08-06 · checked 2026-08-17
Claude Sonnet 4.6 (max) claude-sonnet-4-6-max	Claude Sonnet 4.6	78.8%	1.0.11Epoch AI benchmark runner	Public reference · not scored source-checked	Epoch AI permanent refresh source ↗Observed 2026-08-06 · checked 2026-08-17
Claude Fable 5 (low) claude-fable-5-epoch-claude-fable-5-low	Claude Fable 5	78.8%	1.0.11Epoch AI benchmark runner	Public reference · not scored source-checked	Epoch AI permanent refresh source ↗Observed 2026-08-06 · checked 2026-08-17
Gemini 3.5 Flash-Lite (low) gemini-3-5-flash-lite-epoch-gemini-3-5-flash-lite-low	Gemini 3.5 Flash-Lite	75.8%	1.0.11Epoch AI benchmark runner	Public reference · not scored source-checked	Epoch AI permanent refresh source ↗Observed 2026-08-06 · checked 2026-08-17
Gemma 4 31B IT (minimal) gemma-4-31b-epoch-gemma-4-31b-it-minimal	Gemma 4 31B	75.8%	1.0.11Epoch AI benchmark runner	Public reference · not scored source-checked	Epoch AI permanent refresh source ↗Observed 2026-08-06 · checked 2026-08-17
Gemini 3.5 Flash-Lite (minimal) gemini-3-5-flash-lite-epoch-gemini-3-5-flash-lite-minimal	Gemini 3.5 Flash-Lite	74.2%	1.0.11Epoch AI benchmark runner	Public reference · not scored source-checked	Epoch AI permanent refresh source ↗Observed 2026-08-06 · checked 2026-08-17
Gemini 3.1 Flash-Lite (low) gemini-3-1-flash-lite-epoch-gemini-3-1-flash-lite-low	Gemini 3.1 Flash-Lite	74.2%	1.0.11Epoch AI benchmark runner	Public reference · not scored source-checked	Epoch AI permanent refresh source ↗Observed 2026-08-06 · checked 2026-08-17
Gemini 3.1 Flash-Lite (minimal) gemini-3-1-flash-lite-epoch-gemini-3-1-flash-lite-minimal	Gemini 3.1 Flash-Lite	73.7%	1.0.11Epoch AI benchmark runner	Public reference · not scored source-checked	Epoch AI permanent refresh source ↗Observed 2026-08-06 · checked 2026-08-17
DeepSeek v4 (no thinking) deepseek-v4-pro-epoch-deepseek-v4-pro-none	DeepSeek V4 Pro	73.2%	1.0.11Epoch AI benchmark runner	Public reference · not scored source-checked	Epoch AI permanent refresh source ↗Observed 2026-08-06 · checked 2026-08-18
gpt-oss-20b_high gpt-oss-20b-epoch-gpt-oss-20b-high	GPT-OSS 20B	46.0%	1.0.11Epoch AI benchmark runner	Public reference · not scored source-checked	Epoch AI permanent refresh source ↗Observed 2026-08-06 · checked 2026-08-18
gemma-3-27b-it gemma-3-27b-epoch-gemma-3-27b-it	Gemma 3 27B	43.9%	1.0.11Epoch AI benchmark runner	Public reference · not scored source-checked	Epoch AI permanent refresh source ↗Observed 2026-08-06 · checked 2026-08-18
Inkling (xhigh) inkling-xhigh	Inkling	88.3%	1.0.11Epoch AI benchmark runner	Score input source-checked	Epoch AI permanent refresh source ↗Observed 2026-08-05 · checked 2026-08-18
Qwen3.8 Max (xhigh) qwen-3-8-max-xhigh	Qwen3.8 Max	92.7%	1.0.11Epoch AI benchmark runner	Score input source-checked	Epoch AI permanent refresh source ↗Observed 2026-08-04 · checked 2026-08-17
Qwen3.8 Max (xhigh, default reasoning effort) qwen-3-8-max-xhigh	Qwen3.8 Max	92.6%	CurrentQwen provider evaluation	Public reference · not scored provider-reported	Qwen3.8 Max official release ↗Observed 2026-08-03 · checked 2026-08-05
Gemini 3.6 Flash (high) gemini-3-6-flash-high	Gemini 3.6 Flash	94.1%	1.0.11Epoch AI benchmark runner	Score input source-checked	Epoch AI permanent refresh source ↗Observed 2026-08-02 · checked 2026-08-17
DeepSeek V4 Flash 0731 (max) deepseek-v4-flash-0731-epoch-deepseek-v4-flash-0731-max	DeepSeek V4 Flash 0731	91.0%	1.0.11Epoch AI benchmark runner	Public reference · not scored source-checked	Epoch AI permanent refresh source ↗Observed 2026-08-02 · checked 2026-08-17
Exact BenchLM registry variant Sakana Fugu; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Sakana Fugu	95.5%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Sakana Fugu-Ultra; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Sakana Fugu-Ultra	95.5%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant GPT-5.6 Sol; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.6 Sol	94.6%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Claude Opus 4.7 (Adaptive); bulk export does not retain a complete upstream harness configuration. claude-opus-4-7-max	Claude Opus 4.7	94.2%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Claude Opus 4.8; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude Opus 4.8	93.6%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant GPT-5.5; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.5	93.6%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Kimi K3; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Kimi K3	93.5%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Claude Opus 5; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude Opus 5	93.2%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Grok 4.5; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Grok 4.5	93.1%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant GPT-5.6 Terra; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.6 Terra	92.9%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant MiniMax M3; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	MiniMax M3	92.9%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Gemini 3.6 Flash; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Gemini 3.6 Flash	92.8%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant GPT-5.4; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.4	92.8%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Gemini 3.5 Flash; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Gemini 3.5 Flash	92.7%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Claude Fable 5; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude Fable 5	92.6%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Qwen3.7 Max; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Qwen3.7-Max	92.4%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant GPT-5.6 Luna; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.6 Luna	92.3%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant GPT-5.3 Codex; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.3-Codex	91.5%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant GLM-5.2; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GLM-5.2	91.2%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Claude Sonnet 5; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude Sonnet 5	91.1%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Kimi K2.6; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Kimi K2.6	90.5%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant GPT-5.2; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.2	90.3%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Qwen3.7 Plus; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Qwen3.7-Plus	90.3%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant DeepSeek V4 Pro (Max); bulk export does not retain a complete upstream harness configuration. deepseek-v4-pro-max	DeepSeek V4 Pro	90.1%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Grok 4.3; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Grok 4.3	90.1%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant GPT-5.2-Codex; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.2 Codex	89.9%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Interfaze Beta; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Interfaze Beta	89.9%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Muse Spark 1.1; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Muse Spark 1.1	89.8%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Hy3; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Hy3	89.7%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Claude Opus 4.6 (Adaptive); bulk export does not retain a complete upstream harness configuration. claude-opus-4-6-max	Claude Opus 4.6	89.6%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Kimi K2.7 Code; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Kimi K2.7 Code	89.6%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Inkling-Small; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Inkling-Small	89.5%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Muse Spark; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Muse Spark	89.5%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Qwen3.5 397B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Qwen3.5 397B A17B	89.3%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Qwen3.5 397B (Reasoning); bulk export does not retain a complete upstream harness configuration. qwen3-5-397b-thinking	Qwen3.5 397B A17B	89.3%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Claude Opus 4.6; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude Opus 4.6	89.2%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant DeepSeek V4 Pro (High); bulk export does not retain a complete upstream harness configuration. deepseek-v4-pro-high	DeepSeek V4 Pro	89.1%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Qwen 3.6 Max (preview); bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Qwen 3.6 Max (preview)	88.8%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Claude Opus 4.7; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude Opus 4.7	88.5%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Grok 4.20; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Grok 4.20	88.5%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Qwen3.6 Plus; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Qwen3.6 Plus	88.2%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant DeepSeek V4 Flash (Max); bulk export does not retain a complete upstream harness configuration. deepseek-v4-flash-max	DeepSeek V4 Flash	88.1%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Inkling; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Inkling	87.9%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Kimi K2.5 (Reasoning); bulk export does not retain a complete upstream harness configuration. kimi-k2-5-thinking	Kimi K2.5	87.9%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Grok 4; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Grok 4	87.7%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Kimi K2.5; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Kimi K2.5	87.6%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant GPT-5.4 mini; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.4 mini	87.5%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant DeepSeek V4 Flash (High); bulk export does not retain a complete upstream harness configuration. deepseek-v4-flash-high	DeepSeek V4 Flash	87.4%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant GPT-5.1; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.1	87.3%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Hy3 Preview; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Hy3 Preview	87.2%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant MiMo-V2-Pro; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	MiMo-V2-Pro	87%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant MiniMax M2.7; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	MiniMax M2.7	87%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Nemotron 3 Ultra; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Nemotron 3 Ultra	87%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Claude Opus 4.5 Thinking; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude Opus 4.5 Thinking	86.6%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant MiMo-V2.5-Pro; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	MiMo-V2.5-Pro	86.6%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant GLM-5.1; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GLM-5.1	86.2%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant GLM-5; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GLM-5	86%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant GPT-5.1-Codex; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.1-Codex	86%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant GPT-5.1-Codex-Max; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.1-Codex-Max	86%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant GLM-4.7; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GLM-4.7	85.9%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Qwen3.5-27B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Qwen3.5 27B	85.8%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Gemma 4 31B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Gemma 4 31B	85.7%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Qwen3.5-122B-A10B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Qwen3.5-122B-A10B	85.7%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant GPT-5 (high); bulk export does not retain a complete upstream harness configuration. gpt-5-high	GPT-5	85.4%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Grok 4.1 Fast (Reasoning); bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Grok 4.1 Fast (Reasoning)	85.3%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant GLM-5-Turbo; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GLM-5-Turbo	84.7%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Grok 4 Fast (Reasoning); bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Grok 4 Fast (Reasoning)	84.7%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant o3-pro; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	o3-pro	84.5%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Qwen3.5-35B-A3B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Qwen3.5-35B-A3B	84.5%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Gemini 2.5 Pro; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Gemini 2.5 Pro	84.4%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant GPT-5 (medium); bulk export does not retain a complete upstream harness configuration. gpt-5-medium	GPT-5	84.2%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant MAI-Thinking-1; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	MAI-Thinking-1	84.2%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Qwen3.6-27B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Qwen3.6 27B	84.2%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Qwen3.6-35B-A3B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Qwen3.6-35B-A3B	84.1%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Gemini 3.5 Flash-Lite; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Gemini 3.5 Flash-Lite	83.8%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant MiMo-V2-Omni; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	MiMo-V2-Omni	82.8%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant o3; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	o3	82.7%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Gemini 3.1 Flash-Lite; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Gemini 3.1 Flash-Lite	82.2%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant GPT-5.4 nano; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.4 nano	81.7%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant DeepSeek-R1; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	DeepSeek-R1	81.3%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Gemini 3 Flash; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Gemini 3 Flash	81.2%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Claude Opus 4.5; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude Opus 4.5	81%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Claude 4.1 Opus Thinking; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude 4.1 Opus Thinking	80.9%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant GLM-5V-Turbo; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GLM-5V-Turbo	80.9%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Step 3.7 Flash; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Step 3.7 Flash	80.9%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Claude Sonnet 4.6; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude Sonnet 4.6	79.9%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Gemma 4 26B A4B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Gemma 4 26B A4B	79.2%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Gemma 4 12B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Gemma 4 12B Unified	78.8%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant K-Exaone; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	K-Exaone	78.3%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant GPT-OSS 120B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-OSS 120B	78.2%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant DeepSeek V3.1 (Reasoning); bulk export does not retain a complete upstream harness configuration. deepseek-v3-1-reasoning-default	DeepSeek V3.1	77.9%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Mistral Small 4; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Mistral Small 4	76.9%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Mistral Small 4 (Reasoning); bulk export does not retain a complete upstream harness configuration. mistral-small-4-reasoning	Mistral Small 4	76.9%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Kimi K2; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Kimi K2	76.6%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Qwen3 Max; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Qwen3 Max	76.4%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Trinity-Large-Thinking; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Trinity-Large-Thinking	76.3%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Command A+; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Command A+	76.1%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Nemotron 3 Nano 30B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Nemotron 3 Nano 30B	75.7%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant DeepSeek V3.2; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	DeepSeek V3.2	75.1%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Mistral Medium 3.5 128B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Mistral Medium 3.5 128B	74.8%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant o3-mini; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	o3-mini	74.8%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant o1; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	o1	74.7%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Sarvam 105B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Sarvam 105B	73.8%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant DeepSeek V3.1; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	DeepSeek V3.1	73.5%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant GLM-4.5-Air; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GLM-4.5-Air	73.3%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant DeepSeek V4 Pro; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	DeepSeek V4 Pro	72.9%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Nemotron Ultra 253B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Nemotron Ultra 253B	72.8%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Grok Code Fast 1; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Grok Code Fast 1	72.7%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Nemotron 3 Nano Omni 30B A3B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Nemotron 3 Nano Omni 30B A3B	72.2%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant DeepSeek V4 Flash; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	DeepSeek V4 Flash	71.2%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant ZAYA1-8B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	ZAYA1-8B	71%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant GPT-OSS 20B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-OSS 20B	68.8%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Claude 4 Sonnet; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude 4 Sonnet	68.3%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Gemini 2.5 Flash; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Gemini 2.5 Flash	68.3%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Mistral Large 3; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Mistral Large 3	68%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Llama 4 Maverick; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Llama 4 Maverick	67.1%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant GPT-4.1; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-4.1	66.6%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant GPT-4.1 mini; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-4.1 mini	66.4%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Grok 4.1 Fast; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Grok 4.1 Fast	63.7%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Sarvam 30B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Sarvam 30B	63.3%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Trinity-Large-Preview; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Trinity-Large-Preview	63.3%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant GLM-4.6; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GLM-4.6	63.2%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Exaone 4.0 32B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Exaone 4.0 32B	62.8%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant DeepSeek R1 Distill Qwen 32B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	DeepSeek R1 Distill Qwen 32B	61.5%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Ling 2.6 Flash; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Ling 2.6 Flash	59.3%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Gemini 1.5 Pro; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Gemini 1.5 Pro	58.9%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Llama 4 Scout; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Llama 4 Scout	58.7%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Mistral Medium 3; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Mistral Medium 3	57.8%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Mellum2-12B-A2.5B-Thinking; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Mellum2-12B-A2.5B-Thinking	57.6%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Phi-4; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Phi-4	57.5%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant ZAYA1-74B-Preview; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	ZAYA1-74B-Preview	57.3%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Solar Pro 2; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Solar Pro 2	56.1%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant DeepSeek V3; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	DeepSeek V3	55.7%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant GPT-4o; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-4o	54.3%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Gemma 4 E4B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Gemma 4 E4B	52.2%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Llama 3.1 405B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Llama 3.1 405B	51.5%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant GPT-4.1 nano; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-4.1 nano	51.2%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Nova Pro; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Nova Pro	49.9%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Claude 3 Opus; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude 3 Opus	48.9%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Mistral Large 2; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Mistral Large 2	48.6%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant LFM2.5-8B-A1B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	LFM2.5-8B-A1B	46.6%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Soofi S 30B-A3B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Soofi S 30B-A3B	43.4%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Gemma 3 27B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Gemma 3 27B	42.8%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant GPT-4o mini; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-4o mini	42.6%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Exaone 4.0 1.2B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Exaone 4.0 1.2B	42.4%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Qwen2.5 Coder 32B Instruct; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Qwen2.5 Coder 32B Instruct	41.7%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Mellum2-12B-A2.5B-Instruct; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Mellum2-12B-A2.5B-Instruct	40.9%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Gemma 4 E2B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Gemma 4 E2B	40.5%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Claude 3 Haiku; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude 3 Haiku	37.4%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Granite-4.0-H-350M; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Granite-4.0-H-350M	29%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant LFM2.5-VL-1.6B-Extract; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	LFM2.5-VL-1.6B-Extract	28.9%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Granite-4.0-1B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Granite-4.0-1B	28.1%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Gemini 1.0 Pro; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Gemini 1.0 Pro	27.7%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant MiniCPM5-1B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	MiniCPM5-1B	26.3%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant LFM2.5-230M; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	LFM2.5-230M	25.4%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Granite-4.0-H-1B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Granite-4.0-H-1B	24.6%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Granite-4.0-350M; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Granite-4.0-350M	23.7%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Sakana Fugu; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Sakana Fugu	95.5%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Sakana Fugu-Ultra; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Sakana Fugu-Ultra	95.5%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant GPT-5.6 Sol; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.6 Sol	94.6%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Claude Opus 4.7 (Adaptive); bulk export does not retain a complete upstream harness configuration. claude-opus-4-7-max	Claude Opus 4.7	94.2%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Claude Opus 4.8; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude Opus 4.8	93.6%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant GPT-5.5; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.5	93.6%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Kimi K3; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Kimi K3	93.5%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Claude Opus 5; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude Opus 5	93.2%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Grok 4.5; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Grok 4.5	93.1%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant GPT-5.6 Terra; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.6 Terra	92.9%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant MiniMax M3; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	MiniMax M3	92.9%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Gemini 3.6 Flash; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Gemini 3.6 Flash	92.8%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant GPT-5.4; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.4	92.8%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Gemini 3.5 Flash; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Gemini 3.5 Flash	92.7%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Claude Fable 5; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude Fable 5	92.6%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Qwen3.7 Max; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Qwen3.7-Max	92.4%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant GPT-5.6 Luna; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.6 Luna	92.3%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant GPT-5.3 Codex; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.3-Codex	91.5%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant GLM-5.2; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GLM-5.2	91.2%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Claude Sonnet 5; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude Sonnet 5	91.1%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Kimi K2.6; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Kimi K2.6	90.5%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant GPT-5.2; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.2	90.3%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Qwen3.7 Plus; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Qwen3.7-Plus	90.3%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant DeepSeek V4 Pro (Max); bulk export does not retain a complete upstream harness configuration. deepseek-v4-pro-max	DeepSeek V4 Pro	90.1%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Grok 4.3; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Grok 4.3	90.1%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant GPT-5.2-Codex; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.2 Codex	89.9%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Interfaze Beta; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Interfaze Beta	89.9%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Muse Spark 1.1; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Muse Spark 1.1	89.8%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Hy3; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Hy3	89.7%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Claude Opus 4.6 (Adaptive); bulk export does not retain a complete upstream harness configuration. claude-opus-4-6-max	Claude Opus 4.6	89.6%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Kimi K2.7 Code; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Kimi K2.7 Code	89.6%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Muse Spark; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Muse Spark	89.5%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Qwen3.5 397B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Qwen3.5 397B A17B	89.3%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Qwen3.5 397B (Reasoning); bulk export does not retain a complete upstream harness configuration. qwen3-5-397b-thinking	Qwen3.5 397B A17B	89.3%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Claude Opus 4.6; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude Opus 4.6	89.2%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant DeepSeek V4 Pro (High); bulk export does not retain a complete upstream harness configuration. deepseek-v4-pro-high	DeepSeek V4 Pro	89.1%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Qwen 3.6 Max (preview); bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Qwen 3.6 Max (preview)	88.8%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Claude Opus 4.7; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude Opus 4.7	88.5%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Grok 4.20; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Grok 4.20	88.5%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Qwen3.6 Plus; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Qwen3.6 Plus	88.2%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant DeepSeek V4 Flash (Max); bulk export does not retain a complete upstream harness configuration. deepseek-v4-flash-max	DeepSeek V4 Flash	88.1%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Inkling; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Inkling	87.9%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Kimi K2.5 (Reasoning); bulk export does not retain a complete upstream harness configuration. kimi-k2-5-thinking	Kimi K2.5	87.9%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Grok 4; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Grok 4	87.7%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Kimi K2.5; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Kimi K2.5	87.6%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant GPT-5.4 mini; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.4 mini	87.5%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant DeepSeek V4 Flash (High); bulk export does not retain a complete upstream harness configuration. deepseek-v4-flash-high	DeepSeek V4 Flash	87.4%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant GPT-5.1; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.1	87.3%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Hy3 Preview; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Hy3 Preview	87.2%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant MiMo-V2-Pro; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	MiMo-V2-Pro	87%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant MiniMax M2.7; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	MiniMax M2.7	87%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Nemotron 3 Ultra; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Nemotron 3 Ultra	87%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Claude Opus 4.5 Thinking; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude Opus 4.5 Thinking	86.6%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant MiMo-V2.5-Pro; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	MiMo-V2.5-Pro	86.6%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant GLM-5.1; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GLM-5.1	86.2%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant GLM-5; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GLM-5	86%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant GPT-5.1-Codex; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.1-Codex	86%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant GPT-5.1-Codex-Max; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.1-Codex-Max	86%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant GLM-4.7; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GLM-4.7	85.9%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Qwen3.5-27B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Qwen3.5 27B	85.8%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Gemma 4 31B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Gemma 4 31B	85.7%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Qwen3.5-122B-A10B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Qwen3.5-122B-A10B	85.7%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant GPT-5 (high); bulk export does not retain a complete upstream harness configuration. gpt-5-high	GPT-5	85.4%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Grok 4.1 Fast (Reasoning); bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Grok 4.1 Fast (Reasoning)	85.3%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant GLM-5-Turbo; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GLM-5-Turbo	84.7%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Grok 4 Fast (Reasoning); bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Grok 4 Fast (Reasoning)	84.7%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant o3-pro; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	o3-pro	84.5%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Qwen3.5-35B-A3B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Qwen3.5-35B-A3B	84.5%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Gemini 2.5 Pro; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Gemini 2.5 Pro	84.4%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant GPT-5 (medium); bulk export does not retain a complete upstream harness configuration. gpt-5-medium	GPT-5	84.2%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant MAI-Thinking-1; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	MAI-Thinking-1	84.2%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Qwen3.6-27B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Qwen3.6 27B	84.2%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Qwen3.6-35B-A3B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Qwen3.6-35B-A3B	84.1%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Gemini 3.5 Flash-Lite; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Gemini 3.5 Flash-Lite	83.8%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant MiMo-V2-Omni; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	MiMo-V2-Omni	82.8%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant o3; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	o3	82.7%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Gemini 3.1 Flash-Lite; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Gemini 3.1 Flash-Lite	82.2%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant GPT-5.4 nano; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.4 nano	81.7%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant DeepSeek-R1; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	DeepSeek-R1	81.3%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Gemini 3 Flash; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Gemini 3 Flash	81.2%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Claude Opus 4.5; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude Opus 4.5	81%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Claude 4.1 Opus Thinking; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude 4.1 Opus Thinking	80.9%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant GLM-5V-Turbo; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GLM-5V-Turbo	80.9%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Step 3.7 Flash; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Step 3.7 Flash	80.9%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Claude Sonnet 4.6; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude Sonnet 4.6	79.9%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Gemma 4 26B A4B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Gemma 4 26B A4B	79.2%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Gemma 4 12B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Gemma 4 12B Unified	78.8%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant K-Exaone; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	K-Exaone	78.3%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant GPT-OSS 120B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-OSS 120B	78.2%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant DeepSeek V3.1 (Reasoning); bulk export does not retain a complete upstream harness configuration. deepseek-v3-1-reasoning-default	DeepSeek V3.1	77.9%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Mistral Small 4; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Mistral Small 4	76.9%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Mistral Small 4 (Reasoning); bulk export does not retain a complete upstream harness configuration. mistral-small-4-reasoning	Mistral Small 4	76.9%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Kimi K2; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Kimi K2	76.6%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Qwen3 Max; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Qwen3 Max	76.4%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Trinity-Large-Thinking; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Trinity-Large-Thinking	76.3%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Command A+; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Command A+	76.1%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Nemotron 3 Nano 30B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Nemotron 3 Nano 30B	75.7%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant DeepSeek V3.2; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	DeepSeek V3.2	75.1%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Mistral Medium 3.5 128B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Mistral Medium 3.5 128B	74.8%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant o3-mini; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	o3-mini	74.8%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant o1; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	o1	74.7%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Sarvam 105B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Sarvam 105B	73.8%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant DeepSeek V3.1; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	DeepSeek V3.1	73.5%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant GLM-4.5-Air; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GLM-4.5-Air	73.3%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant DeepSeek V4 Pro; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	DeepSeek V4 Pro	72.9%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Nemotron Ultra 253B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Nemotron Ultra 253B	72.8%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Grok Code Fast 1; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Grok Code Fast 1	72.7%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Nemotron 3 Nano Omni 30B A3B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Nemotron 3 Nano Omni 30B A3B	72.2%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant DeepSeek V4 Flash; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	DeepSeek V4 Flash	71.2%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant ZAYA1-8B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	ZAYA1-8B	71%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant GPT-OSS 20B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-OSS 20B	68.8%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Claude 4 Sonnet; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude 4 Sonnet	68.3%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Gemini 2.5 Flash; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Gemini 2.5 Flash	68.3%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Mistral Large 3; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Mistral Large 3	68%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Llama 4 Maverick; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Llama 4 Maverick	67.1%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant GPT-4.1; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-4.1	66.6%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant GPT-4.1 mini; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-4.1 mini	66.4%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Grok 4.1 Fast; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Grok 4.1 Fast	63.7%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Sarvam 30B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Sarvam 30B	63.3%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Trinity-Large-Preview; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Trinity-Large-Preview	63.3%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant GLM-4.6; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GLM-4.6	63.2%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Exaone 4.0 32B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Exaone 4.0 32B	62.8%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant DeepSeek R1 Distill Qwen 32B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	DeepSeek R1 Distill Qwen 32B	61.5%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Ling 2.6 Flash; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Ling 2.6 Flash	59.3%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Gemini 1.5 Pro; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Gemini 1.5 Pro	58.9%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Llama 4 Scout; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Llama 4 Scout	58.7%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Mistral Medium 3; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Mistral Medium 3	57.8%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Mellum2-12B-A2.5B-Thinking; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Mellum2-12B-A2.5B-Thinking	57.6%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Phi-4; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Phi-4	57.5%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant ZAYA1-74B-Preview; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	ZAYA1-74B-Preview	57.3%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Solar Pro 2; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Solar Pro 2	56.1%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant DeepSeek V3; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	DeepSeek V3	55.7%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant GPT-4o; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-4o	54.3%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Gemma 4 E4B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Gemma 4 E4B	52.2%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Llama 3.1 405B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Llama 3.1 405B	51.5%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant GPT-4.1 nano; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-4.1 nano	51.2%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Nova Pro; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Nova Pro	49.9%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Claude 3 Opus; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude 3 Opus	48.9%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Mistral Large 2; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Mistral Large 2	48.6%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant LFM2.5-8B-A1B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	LFM2.5-8B-A1B	46.6%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Soofi S 30B-A3B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Soofi S 30B-A3B	43.4%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Gemma 3 27B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Gemma 3 27B	42.8%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant GPT-4o mini; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-4o mini	42.6%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Exaone 4.0 1.2B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Exaone 4.0 1.2B	42.4%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Qwen2.5 Coder 32B Instruct; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Qwen2.5 Coder 32B Instruct	41.7%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Mellum2-12B-A2.5B-Instruct; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Mellum2-12B-A2.5B-Instruct	40.9%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Gemma 4 E2B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Gemma 4 E2B	40.5%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Claude 3 Haiku; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude 3 Haiku	37.4%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Granite-4.0-H-350M; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Granite-4.0-H-350M	29%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant LFM2.5-VL-1.6B-Extract; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	LFM2.5-VL-1.6B-Extract	28.9%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Granite-4.0-1B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Granite-4.0-1B	28.1%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Gemini 1.0 Pro; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Gemini 1.0 Pro	27.7%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant MiniCPM5-1B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	MiniCPM5-1B	26.3%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant LFM2.5-230M; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	LFM2.5-230M	25.4%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Granite-4.0-H-1B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Granite-4.0-H-1B	24.6%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Granite-4.0-350M; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Granite-4.0-350M	23.7%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Claude Opus 5 (max) claude-opus-5-max	Claude Opus 5	93.9%	1.0.11Epoch AI benchmark runner	Score input source-checked	Epoch AI permanent refresh source ↗Observed 2026-07-24 · checked 2026-08-17
Exact BenchLM registry variant Sakana Fugu; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Sakana Fugu	95.5%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant Sakana Fugu-Ultra; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Sakana Fugu-Ultra	95.5%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant GPT-5.6 Sol; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.6 Sol	94.6%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant Claude Opus 4.7 (Adaptive); bulk export does not retain a complete upstream harness configuration. claude-opus-4-7-max	Claude Opus 4.7	94.2%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant Claude Opus 4.8; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude Opus 4.8	93.6%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant GPT-5.5; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.5	93.6%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant Kimi K3; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Kimi K3	93.5%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant Grok 4.5; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Grok 4.5	93.1%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant GPT-5.6 Terra; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.6 Terra	92.9%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant MiniMax M3; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	MiniMax M3	92.9%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant Gemini 3.6 Flash; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Gemini 3.6 Flash	92.8%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant GPT-5.4; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.4	92.8%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Artificial Analysis GPQA Diamond independent evaluation. source label without registered configuration ID	Gemini 3.6 Flash	92.8%	CurrentArtificial Analysis evaluation harness	Public reference · not scored source-checked	Gemini 3.6 Flash (Artificial Analysis) ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant Gemini 3.5 Flash; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Gemini 3.5 Flash	92.7%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant Claude Fable 5; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude Fable 5	92.6%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant Qwen3.7 Max; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Qwen3.7-Max	92.4%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant GPT-5.6 Luna; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.6 Luna	92.3%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant GPT-5.3 Codex; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.3-Codex	91.5%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant GLM-5.2; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GLM-5.2	91.2%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant Claude Sonnet 5; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude Sonnet 5	91.1%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant Kimi K2.6; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Kimi K2.6	90.5%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant GPT-5.2; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.2	90.3%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant Qwen3.7 Plus; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Qwen3.7-Plus	90.3%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant DeepSeek V4 Pro (Max); bulk export does not retain a complete upstream harness configuration. deepseek-v4-pro-max	DeepSeek V4 Pro	90.1%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant Grok 4.3; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Grok 4.3	90.1%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant GPT-5.2-Codex; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.2 Codex	89.9%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant Interfaze Beta; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Interfaze Beta	89.9%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant Muse Spark 1.1; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Muse Spark 1.1	89.8%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant Hy3; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Hy3	89.7%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant Claude Opus 4.6 (Adaptive); bulk export does not retain a complete upstream harness configuration. claude-opus-4-6-max	Claude Opus 4.6	89.6%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant Kimi K2.7 Code; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Kimi K2.7 Code	89.6%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant Muse Spark; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Muse Spark	89.5%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant Qwen3.5 397B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Qwen3.5 397B A17B	89.3%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant Qwen3.5 397B (Reasoning); bulk export does not retain a complete upstream harness configuration. qwen3-5-397b-thinking	Qwen3.5 397B A17B	89.3%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant Claude Opus 4.6; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude Opus 4.6	89.2%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant DeepSeek V4 Pro (High); bulk export does not retain a complete upstream harness configuration. deepseek-v4-pro-high	DeepSeek V4 Pro	89.1%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant Qwen 3.6 Max (preview); bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Qwen 3.6 Max (preview)	88.8%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant Claude Opus 4.7; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude Opus 4.7	88.5%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant Grok 4.20; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Grok 4.20	88.5%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant Qwen3.6 Plus; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Qwen3.6 Plus	88.2%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant DeepSeek V4 Flash (Max); bulk export does not retain a complete upstream harness configuration. deepseek-v4-flash-max	DeepSeek V4 Flash	88.1%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant Inkling; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Inkling	87.9%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant Kimi K2.5 (Reasoning); bulk export does not retain a complete upstream harness configuration. kimi-k2-5-thinking	Kimi K2.5	87.9%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant Grok 4; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Grok 4	87.7%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant Kimi K2.5; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Kimi K2.5	87.6%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant GPT-5.4 mini; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.4 mini	87.5%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant DeepSeek V4 Flash (High); bulk export does not retain a complete upstream harness configuration. deepseek-v4-flash-high	DeepSeek V4 Flash	87.4%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant GPT-5.1; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.1	87.3%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant Hy3 Preview; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Hy3 Preview	87.2%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant MiMo-V2-Pro; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	MiMo-V2-Pro	87%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant MiniMax M2.7; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	MiniMax M2.7	87%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant Nemotron 3 Ultra; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Nemotron 3 Ultra	87%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant Claude Opus 4.5 Thinking; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude Opus 4.5 Thinking	86.6%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant MiMo-V2.5-Pro; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	MiMo-V2.5-Pro	86.6%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant GLM-5.1; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GLM-5.1	86.2%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant GLM-5; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GLM-5	86%	2023Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant GPT-5.1-Codex; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.1-Codex	86%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant GPT-5.1-Codex-Max; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.1-Codex-Max	86%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant GLM-4.7; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GLM-4.7	85.9%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant Qwen3.5-27B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Qwen3.5 27B	85.8%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact published model / configuration

Canonical product

Score

Version / system

Evidence

Source / observed

Qwen3.8-Flash-Next (Artificial Analysis independent run)

qwen-3-8-flash-next-aa-unspecified

Qwen3.8-Flash-Next

92.3%

diamondArtificial Analysis GPQA Diamond evaluation

Public reference · not scored

independently-verified

Qwen3.8-Flash-Next individual evaluations ↗Observed 2026-08-29 · checked 2026-08-29

Claude Opus 4.7 (Adaptive Reasoning, Max Effort) (Artificial Analysis independent run)

claude-opus-4-7-max

91.4%

diamondArtificial Analysis GPQA Diamond evaluation

Score input

independently-verified

Claude Opus 4.7 (Adaptive Reasoning, Max Effort) individual evaluations ↗Observed 2026-08-29 · checked 2026-08-29

Claude Opus 4.6 (Adaptive Reasoning, Max Effort) (Artificial Analysis independent run)

claude-opus-4-6-max

89.6%

diamondArtificial Analysis GPQA Diamond evaluation

Public reference · not scored

independently-verified

Claude Opus 4.6 (Adaptive Reasoning, Max Effort) individual evaluations ↗Observed 2026-08-29 · checked 2026-08-29

GLM-5.2 (max) (Artificial Analysis independent run)

glm-5-2-max

89.5%

diamondArtificial Analysis GPQA Diamond evaluation

Score input

independently-verified

GLM-5.2 (max) individual evaluations ↗Observed 2026-08-29 · checked 2026-08-29

Qwen3.5 397B A17B (Reasoning) (Artificial Analysis completed independent run)

qwen3-5-397b-thinking

89.3%

diamondArtificial Analysis GPQA Diamond evaluation

Score input

independently-verified

Qwen3.5 397B A17B (Reasoning) current model and individual-evaluation page ↗Observed 2026-08-29 · checked 2026-08-29

Kimi K2.5 (Reasoning) (Artificial Analysis completed independent run)

kimi-k2-5-thinking

87.9%

diamondArtificial Analysis GPQA Diamond evaluation

Score input

independently-verified

Kimi K2.5 (Reasoning) current model and individual-evaluation page ↗Observed 2026-08-29 · checked 2026-08-29

Qwen3.5 122B A10B (Reasoning) (Artificial Analysis completed independent run)

qwen3-5-122b-a10b-aa-reasoning-default

85.7%

diamondArtificial Analysis GPQA Diamond evaluation

Score input

independently-verified

Qwen3.5 122B A10B (Reasoning) current model and individual-evaluation page ↗Observed 2026-08-29 · checked 2026-08-29

Qwen3.6 35B A3B (Reasoning) (Artificial Analysis completed independent run)

qwen3-6-35b-a3b-aa-reasoning-default

84.1%

diamondArtificial Analysis GPQA Diamond evaluation

Score input

independently-verified

Qwen3.6 35B A3B (Reasoning) current model and individual-evaluation page ↗Observed 2026-08-29 · checked 2026-08-29

Gemma 4 26B A4B (Reasoning) (Artificial Analysis completed independent run)

gemma-4-26b-a4b-aa-reasoning-default

79.2%

diamondArtificial Analysis GPQA Diamond evaluation

Score input

independently-verified

Gemma 4 26B A4B (Reasoning) current model and individual-evaluation page ↗Observed 2026-08-29 · checked 2026-08-29

LongCat 2.0 (Artificial Analysis independent run)

longcat-2-0-default

LongCat-2.0

78.0%

diamondArtificial Analysis GPQA Diamond evaluation

Score input

independently-verified

LongCat 2.0 individual evaluations ↗Observed 2026-08-29 · checked 2026-08-29

DeepSeek V3.1 (Reasoning) (Artificial Analysis independent run)

deepseek-v3-1-reasoning-default

77.9%

diamondArtificial Analysis GPQA Diamond evaluation

Public reference · not scored

independently-verified

DeepSeek V3.1 (Reasoning) individual evaluations ↗Observed 2026-08-29 · checked 2026-08-29

Mistral Small 4 (Reasoning) (Artificial Analysis completed independent run)

mistral-small-4-reasoning

76.9%

diamondArtificial Analysis GPQA Diamond evaluation

Score input

independently-verified

Mistral Small 4 (Reasoning) current model and individual-evaluation page ↗Observed 2026-08-29 · checked 2026-08-29

NVIDIA Nemotron 3 Nano 30B A3B (Reasoning) (Artificial Analysis completed independent run)

nemotron-3-nano-30b-aa-reasoning-default

75.7%

diamondArtificial Analysis GPQA Diamond evaluation

Score input

independently-verified

NVIDIA Nemotron 3 Nano 30B A3B (Reasoning) current model and individual-evaluation page ↗Observed 2026-08-29 · checked 2026-08-29

Trinity Large Thinking (Artificial Analysis completed independent run)

trinity-large-thinking-aa-reasoning-default

75.2%

diamondArtificial Analysis GPQA Diamond evaluation

Score input

independently-verified

Trinity Large Thinking current model and individual-evaluation page ↗Observed 2026-08-29 · checked 2026-08-29

Mistral Medium 3.5 (Artificial Analysis completed independent run)

mistral-medium-3-5-128b-aa-reasoning-default

74.8%

diamondArtificial Analysis GPQA Diamond evaluation

Score input

independently-verified

Mistral Medium 3.5 current model and individual-evaluation page ↗Observed 2026-08-29 · checked 2026-08-29

DeepSeek V3.1 (Non-reasoning) (Artificial Analysis independent run)

deepseek-v3-1-non-reasoning

73.5%

diamondArtificial Analysis GPQA Diamond evaluation

Score input

independently-verified

DeepSeek V3.1 (Non-reasoning) individual evaluations ↗Observed 2026-08-29 · checked 2026-08-29

Llama 4 Maverick (Artificial Analysis completed independent run)

llama-4-maverick-aa-non-reasoning-default

67.1%

diamondArtificial Analysis GPQA Diamond evaluation

Score input

independently-verified

Llama 4 Maverick current model and individual-evaluation page ↗Observed 2026-08-29 · checked 2026-08-29

GPT-4.1 mini (Artificial Analysis completed independent run)

gpt-4-1-mini-epoch-gpt-4-1-mini-2025-04-14

66.4%

diamondArtificial Analysis GPQA Diamond evaluation

Score input

independently-verified

GPT-4.1 mini current model and individual-evaluation page ↗Observed 2026-08-29 · checked 2026-08-29

Llama 4 Scout (Artificial Analysis completed independent run)

llama-4-scout-aa-non-reasoning-default

58.7%

diamondArtificial Analysis GPQA Diamond evaluation

Score input

independently-verified

Llama 4 Scout current model and individual-evaluation page ↗Observed 2026-08-29 · checked 2026-08-29

GPT-4.1 nano (Artificial Analysis completed independent run)

gpt-4-1-nano-epoch-gpt-4-1-nano-2025-04-14

51.2%

diamondArtificial Analysis GPQA Diamond evaluation

Score input

independently-verified

GPT-4.1 nano current model and individual-evaluation page ↗Observed 2026-08-29 · checked 2026-08-29

Gemma 3 27B Instruct (Artificial Analysis completed independent run)

gemma-3-27b-aa-non-reasoning-default

42.8%

diamondArtificial Analysis GPQA Diamond evaluation

Score input

independently-verified

Gemma 3 27B Instruct current model and individual-evaluation page ↗Observed 2026-08-29 · checked 2026-08-29

Grok 4.6 (high; Artificial Analysis independent run)

grok-4-6-high

94.9%

diamondArtificial Analysis GPQA Diamond evaluation

Score input

independently-verified

Grok 4.6 individual evaluations ↗Observed 2026-08-27 · checked 2026-08-27

GPT-5.6 Sol (max; Artificial Analysis independent run)

gpt-5-6-sol-max

94.1%

diamondArtificial Analysis GPQA Diamond evaluation

Score input

independently-verified

GPT-5.6 Sol individual evaluations ↗Observed 2026-08-27 · checked 2026-08-27

GPT-5.5 (xhigh; Artificial Analysis independent run)

gpt-5-5-xhigh

93.5%

diamondArtificial Analysis GPQA Diamond evaluation

Score input

independently-verified

GPT-5.5 individual evaluations ↗Observed 2026-08-27 · checked 2026-08-27

Kimi K3 (max; Artificial Analysis independent run)

kimi-k3-max

93.5%

diamondArtificial Analysis GPQA Diamond evaluation

Score input

independently-verified

Kimi K3 individual evaluations ↗Observed 2026-08-27 · checked 2026-08-27

Claude Opus 5 (max; Artificial Analysis independent run)

claude-opus-5-max

93.2%

diamondArtificial Analysis GPQA Diamond evaluation

Score input

independently-verified

Claude Opus 5 individual evaluations ↗Observed 2026-08-27 · checked 2026-08-27

Grok 4.5 (high; Artificial Analysis independent run)

grok-4-5-aa-2-high

93.1%

diamondArtificial Analysis GPQA Diamond evaluation

Score input

independently-verified

Grok 4.5 individual evaluations ↗Observed 2026-08-27 · checked 2026-08-27

DeepSeek V4 Pro 0813 (max; Artificial Analysis independent run)

deepseek-v4-pro-0813-max

DeepSeek V4 Pro 0813

92.8%

diamondArtificial Analysis GPQA Diamond evaluation

Public reference · not scored

independently-verified

DeepSeek V4 Pro 0813 individual evaluations ↗Observed 2026-08-27 · checked 2026-08-27

Gemini 3.6 Flash (high; Artificial Analysis independent run)

gemini-3-6-flash-high

92.8%

diamondArtificial Analysis GPQA Diamond evaluation

Score input

independently-verified

Gemini 3.6 Flash individual evaluations ↗Observed 2026-08-27 · checked 2026-08-27

Claude Fable 5 (max; Artificial Analysis independent run)

claude-fable-5-max

92.6%

diamondArtificial Analysis GPQA Diamond evaluation

Score input

independently-verified

Claude Fable 5 individual evaluations ↗Observed 2026-08-27 · checked 2026-08-27

GPT-5.6 Terra (max; Artificial Analysis independent run)

gpt-5-6-terra-max

92.5%

diamondArtificial Analysis GPQA Diamond evaluation

Score input

independently-verified

GPT-5.6 Terra individual evaluations ↗Observed 2026-08-27 · checked 2026-08-27

Gemini 3.7 Flash (medium; Artificial Analysis independent run)

gemini-3-7-flash-medium

Gemini 3.7 Flash

92.1%

diamondArtificial Analysis GPQA Diamond evaluation

Score input

independently-verified

Gemini 3.7 Flash individual evaluations ↗Observed 2026-08-27 · checked 2026-08-27

Claude Opus 4.8 (max; Artificial Analysis independent run)

claude-opus-4-8-max

92.0%

diamondArtificial Analysis GPQA Diamond evaluation

Score input

independently-verified

Claude Opus 4.8 individual evaluations ↗Observed 2026-08-27 · checked 2026-08-27

GPT-5.4 (xhigh; Artificial Analysis independent run)

gpt-5-4-xhigh

DeepSeek V4 Flash Vision Exp

92.0%

diamondArtificial Analysis GPQA Diamond evaluation

Score input

independently-verified

GPT-5.4 individual evaluations ↗Observed 2026-08-27 · checked 2026-08-27

GLM-5.3 (max; Artificial Analysis independent run)

glm-5-3-max

GLM-5.3

91.7%

diamondArtificial Analysis GPQA Diamond evaluation

Score input

independently-verified

GLM-5.3 individual evaluations ↗Observed 2026-08-27 · checked 2026-08-27

DeepSeek V4 Flash Vision Exp (max; Artificial Analysis independent run)

deepseek-v4-flash-vision-exp-max-harness

91.3%

diamondArtificial Analysis GPQA Diamond evaluation

Public reference · not scored

independently-verified

DeepSeek V4 Flash Vision Exp individual evaluations ↗Observed 2026-08-27 · checked 2026-08-27

GLM-5.3-Flash (max; Artificial Analysis independent run)

glm-5-3-flash-max

GLM-5.3-Flash

91.2%

diamondArtificial Analysis GPQA Diamond evaluation

Score input

independently-verified

GLM-5.3-Flash individual evaluations ↗Observed 2026-08-27 · checked 2026-08-27

Claude Sonnet 5 (max; Artificial Analysis independent run)

claude-sonnet-5-max

91.1%

diamondArtificial Analysis GPQA Diamond evaluation

Score input

independently-verified

Claude Sonnet 5 individual evaluations ↗Observed 2026-08-27 · checked 2026-08-27

GPT-5.6 Luna (max; Artificial Analysis independent run)

gpt-5-6-luna-max

91.1%

diamondArtificial Analysis GPQA Diamond evaluation

Score input

independently-verified

GPT-5.6 Luna individual evaluations ↗Observed 2026-08-27 · checked 2026-08-27

Muse Spark 1.2 (xhigh; Artificial Analysis independent run)

muse-spark-1-2-xhigh

Muse Spark 1.2

90.4%

diamondArtificial Analysis GPQA Diamond evaluation

Score input

independently-verified

Muse Spark 1.2 individual evaluations ↗Observed 2026-08-27 · checked 2026-08-27

Muse Spark 1.1 (xhigh; Artificial Analysis independent run)

muse-spark-1-1-xhigh

89.8%

diamondArtificial Analysis GPQA Diamond evaluation

Score input

independently-verified

Muse Spark 1.1 individual evaluations ↗Observed 2026-08-27 · checked 2026-08-27

Qwen3.8-Flash-Next (xhigh default thinking configuration)

qwen-3-8-flash-next-xhigh

Qwen3.8-Flash-Next

91.7%

Diamondsystem:qwen3-8-flash-next:gpqa

Score input

source-checked

Qwen3.8-Flash-Next current official launch page ↗Observed 2026-08-26 · checked 2026-08-29

Claude Opus 4.6 (Max) as published in Qwen's Qwen3.8-Flash-Next comparison table

claude-opus-4-6-max

91.3%

Diamondsystem:qwen3-8-flash-next-comparison:cell:language:gpqa:4

Relative comparison · not standard direct

provider-reported

Qwen3.8-Flash-Next launch and official provider evaluations ↗Observed 2026-08-26 · checked 2026-08-26

DeepSeek-V4-Flash-0731 (provider-published configuration) as published in Qwen's Qwen3.8-Flash-Next comparison table

deepseek-v4-flash-0731-deepseek-0813-release-unspecified

DeepSeek V4 Flash 0731

90.8%

Diamondsystem:qwen3-8-flash-next-comparison:cell:language:gpqa:3

Relative comparison · not standard direct

provider-reported

Qwen3.8-Flash-Next launch and official provider evaluations ↗Observed 2026-08-26 · checked 2026-08-26

Qwen3.7-Plus (provider-published configuration) as published in Qwen's Qwen3.8-Flash-Next comparison table

qwen-3-7-plus-unspecified

90.3%

Diamondsystem:qwen3-8-flash-next-comparison:cell:language:gpqa:2

Relative comparison · not standard direct

provider-reported

Qwen3.8-Flash-Next launch and official provider evaluations ↗Observed 2026-08-26 · checked 2026-08-26

Qwen3.8-27B (xhigh default thinking configuration) as published in Qwen's Qwen3.8-Flash-Next comparison table

qwen-3-8-27b-xhigh

Qwen3.8-27B

89.2%

Diamondsystem:qwen3-8-flash-next-comparison:cell:language:gpqa:1

Relative comparison · not standard direct

provider-reported

Qwen3.8-Flash-Next launch and official provider evaluations ↗Observed 2026-08-26 · checked 2026-08-26

GLM-5.3 (max)

glm-5-3-max

GLM-5.3

90.9%

1.0.11Epoch AI benchmark runner

Score input

independently-verified

Epoch AI permanent refresh source ↗Observed 2026-08-24 · checked 2026-09-01

Claude Opus 4.6 Max as published by Qwen

claude-opus-4-6-max

91.3%

DiamondQwen3.8-27B official text table

Relative comparison · not standard direct

provider-reported

Qwen3.8-27B official model card ↗Observed 2026-08-15 · checked 2026-08-15

Qwen3.7-Plus as published by Qwen

qwen-3-7-plus-unspecified

90.3%

DiamondQwen3.8-27B official text table

Relative comparison · not standard direct

provider-reported

Qwen3.8-27B official model card ↗Observed 2026-08-15 · checked 2026-08-15

Qwen3.8-27B (xhigh)

qwen-3-8-27b-xhigh

Qwen3.8-27B

89.2%

DiamondQwen3.8-27B official text table

Score input

provider-reported

Qwen3.8-27B official model card ↗Observed 2026-08-15 · checked 2026-08-15

DeepSeek V4 Flash max as published by Upstage

deepseek-v4-flash-max

88.9%

DiamondSolar Open 2 card English table

Relative comparison · not standard direct

provider-reported

Solar Open 2 250B model card ↗Observed 2026-08-15 · checked 2026-08-15

Qwen3.6-27B as published by Qwen

qwen3-6-27b-default

87.8%

DiamondQwen3.8-27B official text table

Relative comparison · not standard direct

provider-reported

Qwen3.8-27B official model card ↗Observed 2026-08-15 · checked 2026-08-15

Solar Open 2 250B (high)

solar-open2-250b-high

Solar Open 2 250B

86.3%

DiamondSolar Open 2 card English table

Score input

provider-reported

Solar Open 2 250B model card ↗Observed 2026-08-15 · checked 2026-08-15

MiMo-V2.5 as published by Upstage

mimo-v2-5-solar-open2-unspecified

MiMo-V2.5

83%

DiamondSolar Open 2 card English table

Relative comparison · not standard direct

provider-reported

Solar Open 2 250B model card ↗Observed 2026-08-15 · checked 2026-08-15

Mistral Medium 3.5 as published by Upstage

mistral-medium-3-5-high

Mistral Medium 3.5

77.5%

DiamondSolar Open 2 card English table

Relative comparison · not standard direct

provider-reported

Solar Open 2 250B model card ↗Observed 2026-08-15 · checked 2026-08-15

Command A+ as published by Upstage

command-a-plus-solar-open2-unspecified

Solar Open 100B (Reasoning)

75.6%

DiamondSolar Open 2 card English table

Relative comparison · not standard direct

provider-reported

Solar Open 2 250B model card ↗Observed 2026-08-15 · checked 2026-08-15

Solar Open 100B as published by Upstage

solar-open-100b-reasoning-high

66.2%

DiamondSolar Open 2 card English table

Relative comparison · not standard direct

provider-reported

Solar Open 2 250B model card ↗Observed 2026-08-15 · checked 2026-08-15

Gemini 3.7 Flash (high)

gemini-3-7-flash-high

Gemini 3.7 Flash

94.8%

1.0.11Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-08-14 · checked 2026-08-17

Grok 4.6 (xhigh)

grok-4-6-xhigh

93.2%

1.0.11Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-08-14 · checked 2026-08-17

Inkling Small (xhigh)

inkling-small-epoch-inkling-small-xhigh

Inkling-Small

88.5%

1.0.11Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-08-14 · checked 2026-08-18

Grok 4.6 (high)

grok-4-6-high

NVIDIA Nemotron 3.5 Lightning 30B-A3B

94.0%

1.0.11Epoch AI benchmark runner

Score input

source-checked

Epoch AI permanent refresh source ↗Observed 2026-08-12 · checked 2026-08-17

NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16; NVIDIA release evaluation; temperature=1.0; top_p=0.95

nemotron-3-5-lightning-30b-a3b-default

75.4%

CurrentNVIDIA NeMo Gym / NeMo Evaluator SDK consistent release harness

Score input

provider-reported

NVIDIA Nemotron 3.5 Lightning 30B-A3B BF16 model card ↗Observed 2026-08-11 · checked 2026-08-12

MiniMax-M3

minimax-m3-epoch-minimax-m3

90.9%

1.0.11Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-08-10 · checked 2026-08-17

GLM-5.1

glm-5-1-epoch-glm-5-1

89.9%

1.0.11Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-08-10 · checked 2026-08-17

Muse Glimmer-30B; high reasoning; temperature=1.0; top_p=0.95; top_k=64

muse-glimmer-30b-high

Muse Glimmer 30B

83.5%

198-question Diamond setArtificial Analysis GPQA Diamond documented evaluation

Public reference · not scored

provider-reported

Muse Glimmer Evaluation Methodology ↗Observed 2026-08-10 · checked 2026-08-10

GLM-5.2 (no thinking)

glm-5-2-epoch-glm-5-2-none

71.2%

1.0.11Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-08-10 · checked 2026-08-17

Kimi K3 (high)

kimi-k3-epoch-kimi-k3-high

91.9%

1.0.11Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-08-07 · checked 2026-08-17

Qwen3.7 Max

qwen-3-7-max-epoch-qwen3-7-max

90.9%

1.0.11Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-08-07 · checked 2026-08-17

GPT-5.6 Sol (low)

gpt-5-6-sol-low

89.9%

1.0.11Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-08-07 · checked 2026-08-17

Qwen3.7 Plus (thinking)

qwen-3-7-plus-epoch-qwen3-7-plus

87.9%

1.0.11Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-08-07 · checked 2026-08-17

Kimi K2.7 Code

kimi-k2-7-code-epoch-kimi-k2-7-code

87.9%

1.0.11Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-08-07 · checked 2026-08-17

GPT-5.6 Terra (low)

gpt-5-6-terra-low

87.4%

1.0.11Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-08-07 · checked 2026-08-17

Qwen3.6 Max Preview (thinking)

qwen3-6-max-epoch-qwen3-6-max-preview

Qwen3.6 Max

87.4%

1.0.11Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-08-07 · checked 2026-08-17

GPT-5.4 mini (xhigh)

gpt-5-4-mini-xhigh

86.9%

1.0.11Epoch AI benchmark runner

Score input

source-checked

Epoch AI permanent refresh source ↗Observed 2026-08-07 · checked 2026-08-17

Qwen3.5 397B-A17B (no thinking)

qwen3-5-397b-epoch-qwen3-5-397b-a17b-none

86.4%

1.0.11Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-08-07 · checked 2026-08-17

Gemini 3.6 Flash (minimal)

gemini-3-6-flash-epoch-gemini-3-6-flash-minimal

85.9%

1.0.11Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-08-07 · checked 2026-08-17

Qwen3.6 27B (thinking)

qwen3-6-27b-epoch-qwen3-6-27b

85.9%

1.0.11Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-08-07 · checked 2026-08-17

Qwen3.5 397B-A17B (thinking)

qwen3-5-397b-epoch-qwen3-5-397b-a17b

85.9%

1.0.11Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-08-07 · checked 2026-08-17

Kimi K3 (low)

kimi-k3-low

84.8%

1.0.11Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-08-07 · checked 2026-08-17

Qwen3.6 27B (no thinking)

qwen3-6-27b-epoch-qwen3-6-27b-none

84.8%

1.0.11Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-08-07 · checked 2026-08-17

GPT-5.6 Sol (none)

gpt-5-6-sol-epoch-gpt-5-6-sol-none

82.8%

1.0.11Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-08-07 · checked 2026-08-17

Qwen3.7 Flash (thinking)

qwen3-7-flash-epoch-qwen3-7-flash

Qwen3.7 Flash

82.3%

1.0.11Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-08-07 · checked 2026-08-18

GPT-5.6 Luna (low)

gpt-5-6-luna-low

82.3%

1.0.11Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-08-07 · checked 2026-08-17

Qwen3.7 Plus (no thinking)

qwen-3-7-plus-epoch-qwen3-7-plus-none

81.8%

1.0.11Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-08-07 · checked 2026-08-17

o3 (medium)

o3-epoch-o3-2025-04-16-medium

80.8%

1.0.11Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-08-07 · checked 2026-08-17

Qwen3.7 Flash (no thinking)

qwen3-7-flash-epoch-qwen3-7-flash-none

Qwen3.7 Flash

80.8%

1.0.11Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-08-07 · checked 2026-08-18

o4-mini (medium)

o4-mini-epoch-o4-mini-2025-04-16-medium

77.8%

1.0.11Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-08-07 · checked 2026-08-17

GPT-5.6 Terra (none)

gpt-5-6-terra-epoch-gpt-5-6-terra-none

77.3%

1.0.11Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-08-07 · checked 2026-08-17

GPT-5.5 (no thinking)

gpt-5-5-epoch-gpt-5-5-none

77.3%

1.0.11Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-08-07 · checked 2026-08-17

GPT-5.4 nano (low)

gpt-5-4-nano-epoch-gpt-5-4-nano-2026-03-17-low

72.2%

1.0.11Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-08-07 · checked 2026-08-17

GPT-5 mini (medium)

gpt-5-mini-medium

71.7%

1.0.11Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-08-07 · checked 2026-08-17

GPT-5.1 (no thinking)

gpt-5-1-epoch-gpt-5-1-2025-11-13-none

66.7%

1.0.11Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-08-07 · checked 2026-08-17

GPT-5.4 mini (none)

gpt-5-4-mini-epoch-gpt-5-4-mini-2026-03-17-none

64.1%

1.0.11Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-08-07 · checked 2026-08-17

GPT-5.6 Luna (none)

gpt-5-6-luna-epoch-gpt-5-6-luna-none

63.6%

1.0.11Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-08-07 · checked 2026-08-17

GPT-5.4 nano (no thinking)

gpt-5-4-nano-epoch-gpt-5-4-nano-2026-03-17-none

55.6%

1.0.11Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-08-07 · checked 2026-08-17

Gemini 3.1 Pro Preview (high)

gemini-3-1-pro-preview-livebench-2026-06-25-high

94.4%

1.0.11Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-08-06 · checked 2026-08-17

Claude Opus 5

claude-opus-5-epoch-claude-opus-5

92.9%

1.0.11Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-08-06 · checked 2026-08-17

DeepSeek v4 (high)

deepseek-v4-pro-high

90.9%

1.0.11Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-08-06 · checked 2026-08-18

Gemini 3 Flash Preview (high)

gemini-3-flash-epoch-gemini-3-flash-preview-high

89.4%

1.0.11Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-08-06 · checked 2026-08-17

Gemini 3.5 Flash (low)

gemini-3-5-flash-epoch-gemini-3-5-flash-low

88.9%

1.0.11Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-08-06 · checked 2026-08-17

Claude Opus 4.8 (low)

claude-opus-4-8-epoch-claude-opus-4-8-low

88.4%

1.0.11Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-08-06 · checked 2026-08-17

Claude Opus 4.6 (max)

claude-opus-4-6-max

88.4%

1.0.11Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-08-06 · checked 2026-08-17

Claude Opus 5 (low)

claude-opus-5-low

87.9%

1.0.11Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-08-06 · checked 2026-08-17

Gemini 3.6 Flash (low)

gemini-3-6-flash-epoch-gemini-3-6-flash-low

86.4%

1.0.11Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-08-06 · checked 2026-08-17

Claude Opus 4.7 (max)

claude-opus-4-7-max

86.4%

1.0.11Epoch AI benchmark runner

Score input

source-checked

Epoch AI permanent refresh source ↗Observed 2026-08-06 · checked 2026-08-17

Claude Fable 5 (max)

claude-fable-5-max

85.9%

1.0.11Epoch AI benchmark runner

Score input

source-checked

Epoch AI permanent refresh source ↗Observed 2026-08-06 · checked 2026-08-17

Claude Opus 4.8 (no thinking)

claude-opus-4-8-epoch-claude-opus-4-8-none

85.4%

1.0.11Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-08-06 · checked 2026-08-17

Gemini 3.5 Flash-Lite (high)

gemini-3-5-flash-lite-livebench-2026-06-25-high

83.3%

1.0.11Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-08-06 · checked 2026-08-17

Claude Fable 5 (high)

claude-fable-5-high

83.3%

1.0.11Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-08-06 · checked 2026-08-17

Gemini 3.1 Flash-Lite (high)

gemini-3-1-flash-lite-epoch-gemini-3-1-flash-lite-high

81.8%

1.0.11Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-08-06 · checked 2026-08-17

Claude Sonnet 5 (max)

claude-sonnet-5-max

80.3%

1.0.11Epoch AI benchmark runner

Score input

source-checked

Epoch AI permanent refresh source ↗Observed 2026-08-06 · checked 2026-08-17

Claude Sonnet 4.6 (max)

claude-sonnet-4-6-max

78.8%

1.0.11Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-08-06 · checked 2026-08-17

Claude Fable 5 (low)

claude-fable-5-epoch-claude-fable-5-low

78.8%

1.0.11Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-08-06 · checked 2026-08-17

Gemini 3.5 Flash-Lite (low)

gemini-3-5-flash-lite-epoch-gemini-3-5-flash-lite-low

75.8%

1.0.11Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-08-06 · checked 2026-08-17

Gemma 4 31B IT (minimal)

gemma-4-31b-epoch-gemma-4-31b-it-minimal

75.8%

1.0.11Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-08-06 · checked 2026-08-17

Gemini 3.5 Flash-Lite (minimal)

gemini-3-5-flash-lite-epoch-gemini-3-5-flash-lite-minimal

74.2%

1.0.11Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-08-06 · checked 2026-08-17

Gemini 3.1 Flash-Lite (low)

gemini-3-1-flash-lite-epoch-gemini-3-1-flash-lite-low

74.2%

1.0.11Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-08-06 · checked 2026-08-17

Gemini 3.1 Flash-Lite (minimal)

gemini-3-1-flash-lite-epoch-gemini-3-1-flash-lite-minimal

73.7%

1.0.11Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-08-06 · checked 2026-08-17

DeepSeek v4 (no thinking)

deepseek-v4-pro-epoch-deepseek-v4-pro-none

73.2%

1.0.11Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-08-06 · checked 2026-08-18

gpt-oss-20b_high

gpt-oss-20b-epoch-gpt-oss-20b-high

46.0%

1.0.11Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-08-06 · checked 2026-08-18

gemma-3-27b-it

gemma-3-27b-epoch-gemma-3-27b-it

43.9%

1.0.11Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-08-06 · checked 2026-08-18

Inkling (xhigh)

inkling-xhigh

88.3%

1.0.11Epoch AI benchmark runner

Score input

source-checked

Epoch AI permanent refresh source ↗Observed 2026-08-05 · checked 2026-08-18

Qwen3.8 Max (xhigh)

qwen-3-8-max-xhigh

Qwen3.8 Max

92.7%

1.0.11Epoch AI benchmark runner

Score input

source-checked

Epoch AI permanent refresh source ↗Observed 2026-08-04 · checked 2026-08-17

Qwen3.8 Max (xhigh, default reasoning effort)

qwen-3-8-max-xhigh

Qwen3.8 Max

92.6%

CurrentQwen provider evaluation

Public reference · not scored

provider-reported

Qwen3.8 Max official release ↗Observed 2026-08-03 · checked 2026-08-05

Gemini 3.6 Flash (high)

gemini-3-6-flash-high

94.1%

1.0.11Epoch AI benchmark runner

Score input

source-checked

Epoch AI permanent refresh source ↗Observed 2026-08-02 · checked 2026-08-17

DeepSeek V4 Flash 0731 (max)

deepseek-v4-flash-0731-epoch-deepseek-v4-flash-0731-max

DeepSeek V4 Flash 0731

91.0%

1.0.11Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-08-02 · checked 2026-08-17

Exact BenchLM registry variant Sakana Fugu; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

95.5%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Sakana Fugu-Ultra; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

95.5%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant GPT-5.6 Sol; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

94.6%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Claude Opus 4.7 (Adaptive); bulk export does not retain a complete upstream harness configuration.

claude-opus-4-7-max

94.2%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Claude Opus 4.8; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

93.6%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant GPT-5.5; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

93.6%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Kimi K3; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

93.5%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Claude Opus 5; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

93.2%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Grok 4.5; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

93.1%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant GPT-5.6 Terra; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

92.9%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant MiniMax M3; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

92.9%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Gemini 3.6 Flash; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

92.8%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant GPT-5.4; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

92.8%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Gemini 3.5 Flash; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

92.7%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Claude Fable 5; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

92.6%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Qwen3.7 Max; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

92.4%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant GPT-5.6 Luna; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

92.3%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant GPT-5.3 Codex; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

91.5%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant GLM-5.2; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

91.2%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Claude Sonnet 5; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

91.1%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Kimi K2.6; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

90.5%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant GPT-5.2; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

90.3%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Qwen3.7 Plus; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

90.3%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant DeepSeek V4 Pro (Max); bulk export does not retain a complete upstream harness configuration.

deepseek-v4-pro-max

90.1%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Grok 4.3; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

90.1%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant GPT-5.2-Codex; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

89.9%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Interfaze Beta; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

89.9%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Muse Spark 1.1; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

89.8%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Hy3; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

89.7%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Claude Opus 4.6 (Adaptive); bulk export does not retain a complete upstream harness configuration.

claude-opus-4-6-max

89.6%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Kimi K2.7 Code; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

89.6%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Inkling-Small; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

Inkling-Small

89.5%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Muse Spark; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

89.5%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Qwen3.5 397B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

89.3%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Qwen3.5 397B (Reasoning); bulk export does not retain a complete upstream harness configuration.

qwen3-5-397b-thinking

89.3%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Claude Opus 4.6; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

89.2%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant DeepSeek V4 Pro (High); bulk export does not retain a complete upstream harness configuration.

deepseek-v4-pro-high

89.1%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Qwen 3.6 Max (preview); bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

88.8%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Claude Opus 4.7; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

88.5%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Grok 4.20; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

88.5%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Qwen3.6 Plus; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

88.2%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant DeepSeek V4 Flash (Max); bulk export does not retain a complete upstream harness configuration.

deepseek-v4-flash-max

88.1%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Inkling; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

87.9%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Kimi K2.5 (Reasoning); bulk export does not retain a complete upstream harness configuration.

kimi-k2-5-thinking

87.9%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Grok 4; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

87.7%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Kimi K2.5; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

87.6%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant GPT-5.4 mini; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

87.5%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant DeepSeek V4 Flash (High); bulk export does not retain a complete upstream harness configuration.

deepseek-v4-flash-high

87.4%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant GPT-5.1; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

87.3%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Hy3 Preview; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

87.2%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant MiMo-V2-Pro; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

87%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant MiniMax M2.7; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

87%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Nemotron 3 Ultra; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

87%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Claude Opus 4.5 Thinking; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

86.6%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant MiMo-V2.5-Pro; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

86.6%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant GLM-5.1; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

86.2%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant GLM-5; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

86%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant GPT-5.1-Codex; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

86%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant GPT-5.1-Codex-Max; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

86%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant GLM-4.7; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

85.9%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Qwen3.5-27B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

85.8%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Gemma 4 31B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

85.7%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Qwen3.5-122B-A10B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

85.7%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant GPT-5 (high); bulk export does not retain a complete upstream harness configuration.

gpt-5-high

85.4%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Grok 4.1 Fast (Reasoning); bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

85.3%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant GLM-5-Turbo; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

84.7%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Grok 4 Fast (Reasoning); bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

84.7%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant o3-pro; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

84.5%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Qwen3.5-35B-A3B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

84.5%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Gemini 2.5 Pro; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

84.4%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant GPT-5 (medium); bulk export does not retain a complete upstream harness configuration.

gpt-5-medium

84.2%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant MAI-Thinking-1; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

84.2%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Qwen3.6-27B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

84.2%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Qwen3.6-35B-A3B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

84.1%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Gemini 3.5 Flash-Lite; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

83.8%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant MiMo-V2-Omni; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

82.8%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant o3; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

82.7%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Gemini 3.1 Flash-Lite; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

82.2%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant GPT-5.4 nano; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

81.7%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant DeepSeek-R1; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

81.3%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Gemini 3 Flash; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

81.2%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Claude Opus 4.5; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

81%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Claude 4.1 Opus Thinking; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

80.9%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant GLM-5V-Turbo; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

80.9%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Step 3.7 Flash; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

80.9%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Claude Sonnet 4.6; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

79.9%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Gemma 4 26B A4B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

79.2%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Gemma 4 12B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

78.8%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant K-Exaone; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

78.3%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant GPT-OSS 120B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

78.2%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant DeepSeek V3.1 (Reasoning); bulk export does not retain a complete upstream harness configuration.

deepseek-v3-1-reasoning-default

77.9%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Mistral Small 4; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

76.9%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Mistral Small 4 (Reasoning); bulk export does not retain a complete upstream harness configuration.

mistral-small-4-reasoning

76.9%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Kimi K2; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

76.6%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Qwen3 Max; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

76.4%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Trinity-Large-Thinking; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

76.3%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Command A+; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

76.1%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Nemotron 3 Nano 30B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

75.7%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant DeepSeek V3.2; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

75.1%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Mistral Medium 3.5 128B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

74.8%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant o3-mini; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

74.8%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant o1; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

74.7%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Sarvam 105B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

73.8%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant DeepSeek V3.1; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

73.5%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant GLM-4.5-Air; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

73.3%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant DeepSeek V4 Pro; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

72.9%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Nemotron Ultra 253B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

72.8%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Grok Code Fast 1; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

72.7%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Nemotron 3 Nano Omni 30B A3B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

72.2%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant DeepSeek V4 Flash; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

71.2%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant ZAYA1-8B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

71%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant GPT-OSS 20B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

68.8%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Claude 4 Sonnet; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

68.3%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Gemini 2.5 Flash; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

68.3%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Mistral Large 3; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

68%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Llama 4 Maverick; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

67.1%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant GPT-4.1; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

66.6%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant GPT-4.1 mini; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

66.4%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Grok 4.1 Fast; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

63.7%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Sarvam 30B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

63.3%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Trinity-Large-Preview; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

63.3%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant GLM-4.6; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

63.2%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Exaone 4.0 32B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

62.8%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant DeepSeek R1 Distill Qwen 32B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

61.5%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Ling 2.6 Flash; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

59.3%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Gemini 1.5 Pro; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

58.9%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Llama 4 Scout; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

58.7%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Mistral Medium 3; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

57.8%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Mellum2-12B-A2.5B-Thinking; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

57.6%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Phi-4; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

57.5%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant ZAYA1-74B-Preview; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

57.3%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Solar Pro 2; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

56.1%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant DeepSeek V3; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

55.7%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant GPT-4o; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

54.3%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Gemma 4 E4B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

52.2%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Llama 3.1 405B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

51.5%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant GPT-4.1 nano; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

51.2%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Nova Pro; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

49.9%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Claude 3 Opus; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

48.9%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Mistral Large 2; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

48.6%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant LFM2.5-8B-A1B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

46.6%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Soofi S 30B-A3B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

43.4%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Gemma 3 27B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

42.8%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant GPT-4o mini; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

42.6%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Exaone 4.0 1.2B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

42.4%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Qwen2.5 Coder 32B Instruct; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

41.7%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Mellum2-12B-A2.5B-Instruct; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

40.9%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Gemma 4 E2B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

40.5%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Claude 3 Haiku; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

37.4%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Granite-4.0-H-350M; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

29%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant LFM2.5-VL-1.6B-Extract; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

28.9%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Granite-4.0-1B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

28.1%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Gemini 1.0 Pro; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

27.7%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant MiniCPM5-1B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

26.3%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant LFM2.5-230M; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

25.4%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Granite-4.0-H-1B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

24.6%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Granite-4.0-350M; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

23.7%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01

Exact BenchLM registry variant Sakana Fugu; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

95.5%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Sakana Fugu-Ultra; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

95.5%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant GPT-5.6 Sol; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

94.6%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Claude Opus 4.7 (Adaptive); bulk export does not retain a complete upstream harness configuration.

claude-opus-4-7-max

94.2%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Claude Opus 4.8; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

93.6%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant GPT-5.5; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

93.6%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Kimi K3; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

93.5%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Claude Opus 5; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

93.2%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Grok 4.5; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

93.1%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant GPT-5.6 Terra; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

92.9%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant MiniMax M3; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

92.9%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Gemini 3.6 Flash; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

92.8%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant GPT-5.4; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

92.8%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Gemini 3.5 Flash; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

92.7%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Claude Fable 5; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

92.6%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Qwen3.7 Max; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

92.4%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant GPT-5.6 Luna; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

92.3%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant GPT-5.3 Codex; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

91.5%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant GLM-5.2; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

91.2%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Claude Sonnet 5; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

91.1%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Kimi K2.6; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

90.5%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant GPT-5.2; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

90.3%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Qwen3.7 Plus; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

90.3%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant DeepSeek V4 Pro (Max); bulk export does not retain a complete upstream harness configuration.

deepseek-v4-pro-max

90.1%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Grok 4.3; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

90.1%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant GPT-5.2-Codex; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

89.9%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Interfaze Beta; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

89.9%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Muse Spark 1.1; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

89.8%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Hy3; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

89.7%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Claude Opus 4.6 (Adaptive); bulk export does not retain a complete upstream harness configuration.

claude-opus-4-6-max

89.6%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Kimi K2.7 Code; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

89.6%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Muse Spark; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

89.5%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Qwen3.5 397B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

89.3%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Qwen3.5 397B (Reasoning); bulk export does not retain a complete upstream harness configuration.

qwen3-5-397b-thinking

89.3%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Claude Opus 4.6; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

89.2%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant DeepSeek V4 Pro (High); bulk export does not retain a complete upstream harness configuration.

deepseek-v4-pro-high

89.1%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Qwen 3.6 Max (preview); bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

88.8%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Claude Opus 4.7; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

88.5%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Grok 4.20; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

88.5%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Qwen3.6 Plus; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

88.2%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant DeepSeek V4 Flash (Max); bulk export does not retain a complete upstream harness configuration.

deepseek-v4-flash-max

88.1%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Inkling; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

87.9%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Kimi K2.5 (Reasoning); bulk export does not retain a complete upstream harness configuration.

kimi-k2-5-thinking

87.9%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Grok 4; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

87.7%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Kimi K2.5; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

87.6%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant GPT-5.4 mini; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

87.5%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant DeepSeek V4 Flash (High); bulk export does not retain a complete upstream harness configuration.

deepseek-v4-flash-high

87.4%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant GPT-5.1; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

87.3%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Hy3 Preview; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

87.2%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant MiMo-V2-Pro; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

87%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant MiniMax M2.7; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

87%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Nemotron 3 Ultra; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

87%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Claude Opus 4.5 Thinking; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

86.6%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant MiMo-V2.5-Pro; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

86.6%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant GLM-5.1; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

86.2%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant GLM-5; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

86%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant GPT-5.1-Codex; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

86%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant GPT-5.1-Codex-Max; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

86%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant GLM-4.7; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

85.9%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Qwen3.5-27B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

85.8%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Gemma 4 31B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

85.7%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Qwen3.5-122B-A10B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

85.7%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant GPT-5 (high); bulk export does not retain a complete upstream harness configuration.

gpt-5-high

85.4%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Grok 4.1 Fast (Reasoning); bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

85.3%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant GLM-5-Turbo; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

84.7%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Grok 4 Fast (Reasoning); bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

84.7%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant o3-pro; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

84.5%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Qwen3.5-35B-A3B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

84.5%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Gemini 2.5 Pro; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

84.4%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant GPT-5 (medium); bulk export does not retain a complete upstream harness configuration.

gpt-5-medium

84.2%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant MAI-Thinking-1; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

84.2%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Qwen3.6-27B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

84.2%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Qwen3.6-35B-A3B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

84.1%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Gemini 3.5 Flash-Lite; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

83.8%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant MiMo-V2-Omni; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

82.8%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant o3; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

82.7%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Gemini 3.1 Flash-Lite; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

82.2%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant GPT-5.4 nano; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

81.7%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant DeepSeek-R1; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

81.3%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Gemini 3 Flash; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

81.2%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Claude Opus 4.5; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

81%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Claude 4.1 Opus Thinking; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

80.9%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant GLM-5V-Turbo; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

80.9%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Step 3.7 Flash; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

80.9%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Claude Sonnet 4.6; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

79.9%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Gemma 4 26B A4B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

79.2%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Gemma 4 12B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

78.8%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant K-Exaone; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

78.3%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant GPT-OSS 120B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

78.2%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant DeepSeek V3.1 (Reasoning); bulk export does not retain a complete upstream harness configuration.

deepseek-v3-1-reasoning-default

77.9%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Mistral Small 4; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

76.9%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Mistral Small 4 (Reasoning); bulk export does not retain a complete upstream harness configuration.

mistral-small-4-reasoning

76.9%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Kimi K2; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

76.6%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Qwen3 Max; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

76.4%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Trinity-Large-Thinking; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

76.3%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Command A+; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

76.1%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Nemotron 3 Nano 30B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

75.7%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant DeepSeek V3.2; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

75.1%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Mistral Medium 3.5 128B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

74.8%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant o3-mini; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

74.8%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant o1; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

74.7%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Sarvam 105B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

73.8%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant DeepSeek V3.1; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

73.5%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant GLM-4.5-Air; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

73.3%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant DeepSeek V4 Pro; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

72.9%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Nemotron Ultra 253B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

72.8%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Grok Code Fast 1; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

72.7%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Nemotron 3 Nano Omni 30B A3B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

72.2%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant DeepSeek V4 Flash; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

71.2%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant ZAYA1-8B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

71%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant GPT-OSS 20B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

68.8%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Claude 4 Sonnet; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

68.3%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Gemini 2.5 Flash; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

68.3%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Mistral Large 3; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

68%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Llama 4 Maverick; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

67.1%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant GPT-4.1; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

66.6%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant GPT-4.1 mini; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

66.4%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Grok 4.1 Fast; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

63.7%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Sarvam 30B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

63.3%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Trinity-Large-Preview; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

63.3%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant GLM-4.6; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

63.2%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Exaone 4.0 32B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

62.8%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant DeepSeek R1 Distill Qwen 32B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

61.5%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Ling 2.6 Flash; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

59.3%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Gemini 1.5 Pro; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

58.9%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Llama 4 Scout; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

58.7%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Mistral Medium 3; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

57.8%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Mellum2-12B-A2.5B-Thinking; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

57.6%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Phi-4; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

57.5%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant ZAYA1-74B-Preview; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

57.3%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Solar Pro 2; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

56.1%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant DeepSeek V3; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

55.7%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant GPT-4o; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

54.3%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Gemma 4 E4B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

52.2%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Llama 3.1 405B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

51.5%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant GPT-4.1 nano; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

51.2%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Nova Pro; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

49.9%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Claude 3 Opus; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

48.9%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Mistral Large 2; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

48.6%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant LFM2.5-8B-A1B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

46.6%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Soofi S 30B-A3B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

43.4%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Gemma 3 27B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

42.8%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant GPT-4o mini; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

42.6%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Exaone 4.0 1.2B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

42.4%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Qwen2.5 Coder 32B Instruct; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

41.7%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Mellum2-12B-A2.5B-Instruct; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

40.9%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Gemma 4 E2B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

40.5%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Claude 3 Haiku; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

37.4%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Granite-4.0-H-350M; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

29%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant LFM2.5-VL-1.6B-Extract; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

28.9%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Granite-4.0-1B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

28.1%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Gemini 1.0 Pro; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

27.7%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant MiniCPM5-1B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

26.3%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant LFM2.5-230M; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

25.4%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Granite-4.0-H-1B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

24.6%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Exact BenchLM registry variant Granite-4.0-350M; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

23.7%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Claude Opus 5 (max)

claude-opus-5-max

93.9%

1.0.11Epoch AI benchmark runner

Score input

source-checked

Epoch AI permanent refresh source ↗Observed 2026-07-24 · checked 2026-08-17

Exact BenchLM registry variant Sakana Fugu; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

95.5%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Sakana Fugu-Ultra; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

95.5%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant GPT-5.6 Sol; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

94.6%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Claude Opus 4.7 (Adaptive); bulk export does not retain a complete upstream harness configuration.

claude-opus-4-7-max

94.2%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Claude Opus 4.8; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

93.6%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant GPT-5.5; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

93.6%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Kimi K3; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

93.5%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Grok 4.5; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

93.1%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant GPT-5.6 Terra; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

92.9%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant MiniMax M3; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

92.9%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Gemini 3.6 Flash; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

92.8%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant GPT-5.4; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

92.8%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Artificial Analysis GPQA Diamond independent evaluation.

source label without registered configuration ID

92.8%

CurrentArtificial Analysis evaluation harness

Public reference · not scored

source-checked

Gemini 3.6 Flash (Artificial Analysis) ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Gemini 3.5 Flash; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

92.7%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Claude Fable 5; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

92.6%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Qwen3.7 Max; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

92.4%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant GPT-5.6 Luna; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

92.3%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant GPT-5.3 Codex; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

91.5%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant GLM-5.2; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

91.2%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Claude Sonnet 5; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

91.1%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Kimi K2.6; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

90.5%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant GPT-5.2; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

90.3%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Qwen3.7 Plus; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

90.3%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant DeepSeek V4 Pro (Max); bulk export does not retain a complete upstream harness configuration.

deepseek-v4-pro-max

90.1%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Grok 4.3; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

90.1%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant GPT-5.2-Codex; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

89.9%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Interfaze Beta; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

89.9%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Muse Spark 1.1; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

89.8%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Hy3; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

89.7%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Claude Opus 4.6 (Adaptive); bulk export does not retain a complete upstream harness configuration.

claude-opus-4-6-max

89.6%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Kimi K2.7 Code; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

89.6%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Muse Spark; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

89.5%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Qwen3.5 397B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

89.3%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Qwen3.5 397B (Reasoning); bulk export does not retain a complete upstream harness configuration.

qwen3-5-397b-thinking

89.3%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Claude Opus 4.6; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

89.2%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant DeepSeek V4 Pro (High); bulk export does not retain a complete upstream harness configuration.

deepseek-v4-pro-high

89.1%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Qwen 3.6 Max (preview); bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

88.8%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Claude Opus 4.7; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

88.5%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Grok 4.20; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

88.5%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Qwen3.6 Plus; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

88.2%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant DeepSeek V4 Flash (Max); bulk export does not retain a complete upstream harness configuration.

deepseek-v4-flash-max

88.1%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Inkling; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

87.9%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Kimi K2.5 (Reasoning); bulk export does not retain a complete upstream harness configuration.

kimi-k2-5-thinking

87.9%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Grok 4; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

87.7%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Kimi K2.5; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

87.6%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant GPT-5.4 mini; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

87.5%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant DeepSeek V4 Flash (High); bulk export does not retain a complete upstream harness configuration.

deepseek-v4-flash-high

87.4%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant GPT-5.1; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

87.3%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Hy3 Preview; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

87.2%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant MiMo-V2-Pro; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

87%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant MiniMax M2.7; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

87%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Nemotron 3 Ultra; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

87%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Claude Opus 4.5 Thinking; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

86.6%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant MiMo-V2.5-Pro; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

86.6%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant GLM-5.1; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

86.2%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant GLM-5; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

86%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant GPT-5.1-Codex; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

86%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant GPT-5.1-Codex-Max; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

86%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant GLM-4.7; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

85.9%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Qwen3.5-27B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

85.8%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Gemma 4 31B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

85.7%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Qwen3.5-122B-A10B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

85.7%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant GPT-5 (high); bulk export does not retain a complete upstream harness configuration.

gpt-5-high

85.4%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Grok 4.1 Fast (Reasoning); bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

85.3%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant GLM-5-Turbo; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

84.7%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Grok 4 Fast (Reasoning); bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

84.7%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant o3-pro; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

84.5%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Qwen3.5-35B-A3B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

84.5%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Gemini 2.5 Pro; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

84.4%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant GPT-5 (medium); bulk export does not retain a complete upstream harness configuration.

gpt-5-medium

84.2%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant MAI-Thinking-1; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

84.2%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Qwen3.6-27B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

84.2%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Qwen3.6-35B-A3B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

84.1%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Gemini 3.5 Flash-Lite; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

83.8%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Artificial Analysis GPQA Diamond independent evaluation.

source label without registered configuration ID

83.8%

CurrentArtificial Analysis evaluation harness

Public reference · not scored

source-checked

Gemini 3.5 Flash-Lite (Artificial Analysis) ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant MiMo-V2-Omni; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

82.8%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant o3; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

82.7%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Gemini 3.1 Flash-Lite; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

82.2%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant GPT-5.4 nano; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

81.7%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant DeepSeek-R1; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

81.3%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Gemini 3 Flash; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

81.2%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Claude Opus 4.5; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

81%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Claude 4.1 Opus Thinking; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

80.9%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant GLM-5V-Turbo; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

80.9%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Step 3.7 Flash; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

80.9%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Claude Sonnet 4.6; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

79.9%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Gemma 4 26B A4B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

79.2%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Gemma 4 12B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

78.8%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant K-Exaone; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

78.3%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant GPT-OSS 120B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

78.2%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant DeepSeek V3.1 (Reasoning); bulk export does not retain a complete upstream harness configuration.

deepseek-v3-1-reasoning-default

77.9%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Mistral Small 4; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

76.9%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Mistral Small 4 (Reasoning); bulk export does not retain a complete upstream harness configuration.

mistral-small-4-reasoning

76.9%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Kimi K2; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

76.6%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Qwen3 Max; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

76.4%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Trinity-Large-Thinking; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

76.3%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Command A+; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

76.1%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Nemotron 3 Nano 30B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

75.7%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant DeepSeek V3.2; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

75.1%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Mistral Medium 3.5 128B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

74.8%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant o3-mini; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

74.8%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant o1; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

74.7%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Sarvam 105B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

73.8%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant DeepSeek V3.1; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

73.5%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant GLM-4.5-Air; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

73.3%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant DeepSeek V4 Pro; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

72.9%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Nemotron Ultra 253B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

72.8%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Grok Code Fast 1; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

72.7%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Nemotron 3 Nano Omni 30B A3B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

72.2%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant DeepSeek V4 Flash; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

71.2%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant ZAYA1-8B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

71%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant GPT-OSS 20B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

68.8%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Claude 4 Sonnet; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

68.3%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Gemini 2.5 Flash; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

68.3%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Mistral Large 3; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

68%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Llama 4 Maverick; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

67.1%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant GPT-4.1; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

66.6%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant GPT-4.1 mini; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

66.4%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Grok 4.1 Fast; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

63.7%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Sarvam 30B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

63.3%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Trinity-Large-Preview; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

63.3%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant GLM-4.6; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

63.2%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Exaone 4.0 32B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

62.8%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant DeepSeek R1 Distill Qwen 32B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

61.5%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Ling 2.6 Flash; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

59.3%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Gemini 1.5 Pro; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

58.9%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Llama 4 Scout; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

58.7%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Mistral Medium 3; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

57.8%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Gemma 4 E4B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

57.6%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Mellum2-12B-A2.5B-Thinking; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

57.6%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Phi-4; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

57.5%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant ZAYA1-74B-Preview; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

57.3%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Solar Pro 2; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

56.1%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant DeepSeek V3; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

55.7%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant GPT-4o; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

54.3%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Llama 3.1 405B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

51.5%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant LFM2.5-8B-A1B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

51.3%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant GPT-4.1 nano; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

51.2%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Nova Pro; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

49.9%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Claude 3 Opus; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

48.9%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Mistral Large 2; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

48.6%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Soofi S 30B-A3B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

43.4%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Gemma 3 27B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

42.8%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant GPT-4o mini; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

42.6%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Exaone 4.0 1.2B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

42.4%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Qwen2.5 Coder 32B Instruct; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

41.7%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Mellum2-12B-A2.5B-Instruct; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

40.9%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Gemma 4 E2B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

37.5%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Claude 3 Haiku; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

37.4%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant LFM2.5-VL-1.6B-Extract; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

28.9%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Granite-4.0-1B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

28.1%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Gemini 1.0 Pro; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

27.7%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Granite-4.0-H-1B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

26.3%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant MiniCPM5-1B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

26.3%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Granite-4.0-H-350M; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

25.7%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant LFM2.5-230M; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

25.4%

2023Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

Exact BenchLM registry variant Granite-4.0-350M; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

20.3%

2026Source-native system

Source carrier · no additional scoring weight

source-checked

BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

o1 (low)

o1-epoch-o1-2024-12-17-low

74.2%

1.0.11Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-07-20 · checked 2026-08-17

GPT-5 (minimal)

gpt-5-epoch-gpt-5-2025-08-07-minimal

71.7%

1.0.11Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-07-20 · checked 2026-08-17

o3-mini-2025-01-31_low

o3-mini-epoch-o3-mini-2025-01-31-low

68.2%

1.0.11Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-07-20 · checked 2026-08-18

GPQA Diamond; max reasoning; provider-published Kimi K3 evaluation.

kimi-k3-max

93.5%

CurrentProvider evaluation harness

Public reference · not scored

provider-reported

Kimi K3: Open Frontier Intelligence ↗Observed 2026-07-16 · checked 2026-07-21

Kimi K3 (Max)

kimi-k3-max

93.1%

1.0.11Epoch AI benchmark runner

Score input

source-checked

Epoch AI permanent refresh source ↗Observed 2026-07-16 · checked 2026-08-17

Muse Spark 1.1 (xhigh); Artificial Analysis independent evaluation configuration as published.

muse-spark-1-1-xhigh

89.8%

CurrentArtificial Analysis independent evaluation

Public reference · not scored

source-checked

Muse Spark 1.1 (xhigh) analysis ↗Observed 2026-07-16 · checked 2026-07-16

deepseek-chat

deepseek-v3-2-epoch-deepseek-chat

71.2%

1.0.11Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-07-16 · checked 2026-08-17

Exact model variant as listed on the BenchLM public benchmark leaderboard page.

source label without registered configuration ID

94.6%

CurrentBenchLM aggregated public evaluation

Source carrier · no additional scoring weight

source-checked

BenchLM public benchmark leaderboards ↗Observed 2026-07-15 · checked 2026-07-15

Artificial Analysis public model evaluation page; exact provider variant as listed on the page.

source label without registered configuration ID

94.1%

CurrentArtificial Analysis evaluation harness

Public reference · not scored

source-checked

Artificial Analysis evaluations for gpt-5-6-sol ↗Observed 2026-07-15 · checked 2026-07-15

Artificial Analysis public model evaluation page; exact provider variant as listed on the page.

source label without registered configuration ID

94.1%

CurrentArtificial Analysis evaluation harness

Public reference · not scored

source-checked

Artificial Analysis evaluations for gemini-3-1-pro-preview ↗Observed 2026-07-15 · checked 2026-07-15

Exact model variant as listed on the BenchLM public benchmark leaderboard page.

source label without registered configuration ID

Claude Mythos 5

94.1%

CurrentBenchLM aggregated public evaluation

Source carrier · no additional scoring weight

source-checked

BenchLM public benchmark leaderboards ↗Observed 2026-07-15 · checked 2026-07-15

Exact model variant as listed on the BenchLM public benchmark leaderboard page.

source label without registered configuration ID

93.6%

CurrentBenchLM aggregated public evaluation

Source carrier · no additional scoring weight

source-checked

BenchLM public benchmark leaderboards ↗Observed 2026-07-15 · checked 2026-07-15

Exact model variant as listed on the BenchLM public benchmark leaderboard page.

source label without registered configuration ID

93.6%

CurrentBenchLM aggregated public evaluation

Source carrier · no additional scoring weight

source-checked

BenchLM public benchmark leaderboards ↗Observed 2026-07-15 · checked 2026-07-15

Artificial Analysis public model evaluation page; exact provider variant as listed on the page.

source label without registered configuration ID

93.5%

CurrentArtificial Analysis evaluation harness

Public reference · not scored

source-checked

Artificial Analysis evaluations for gpt-5-5 ↗Observed 2026-07-15 · checked 2026-07-15

Artificial Analysis public model evaluation page; exact provider variant as listed on the page.

source label without registered configuration ID

93.1%

CurrentArtificial Analysis evaluation harness

Public reference · not scored

source-checked

Artificial Analysis evaluations for grok-4-5 ↗Observed 2026-07-15 · checked 2026-07-15

Artificial Analysis public model evaluation page; exact provider variant as listed on the page.

source label without registered configuration ID

92.9%

CurrentArtificial Analysis evaluation harness

Public reference · not scored

source-checked

Artificial Analysis evaluations for minimax-m3 ↗Observed 2026-07-15 · checked 2026-07-15

Exact model variant as listed on the BenchLM public benchmark leaderboard page.

source label without registered configuration ID

92.9%

CurrentBenchLM aggregated public evaluation

Source carrier · no additional scoring weight

source-checked

BenchLM public benchmark leaderboards ↗Observed 2026-07-15 · checked 2026-07-15

Exact model variant as listed on the BenchLM public benchmark leaderboard page.

source label without registered configuration ID

92.8%

CurrentBenchLM aggregated public evaluation

Source carrier · no additional scoring weight

source-checked

BenchLM public benchmark leaderboards ↗Observed 2026-07-15 · checked 2026-07-15

Artificial Analysis public model evaluation page; exact provider variant as listed on the page.

source label without registered configuration ID

92.6%

CurrentArtificial Analysis evaluation harness

Public reference · not scored

source-checked

Artificial Analysis evaluations for claude-fable-5 ↗Observed 2026-07-15 · checked 2026-07-15

Artificial Analysis public model evaluation page; exact provider variant as listed on the page.

source label without registered configuration ID

92.5%

CurrentArtificial Analysis evaluation harness

Public reference · not scored

source-checked

Artificial Analysis evaluations for gpt-5-6-terra ↗Observed 2026-07-15 · checked 2026-07-15

Exact model variant as listed on the BenchLM public benchmark leaderboard page.

source label without registered configuration ID

92.4%

CurrentBenchLM aggregated public evaluation

Source carrier · no additional scoring weight

source-checked

BenchLM public benchmark leaderboards ↗Observed 2026-07-15 · checked 2026-07-15

Artificial Analysis public model evaluation page; exact provider variant as listed on the page.

source label without registered configuration ID

92.3%

CurrentArtificial Analysis evaluation harness

Public reference · not scored

source-checked

Artificial Analysis evaluations for qwen3-7-max ↗Observed 2026-07-15 · checked 2026-07-15

Exact model variant as listed on the BenchLM public benchmark leaderboard page.

source label without registered configuration ID

92.3%

CurrentBenchLM aggregated public evaluation

Source carrier · no additional scoring weight

source-checked

BenchLM public benchmark leaderboards ↗Observed 2026-07-15 · checked 2026-07-15

Exact model variant as listed on the BenchLM public benchmark leaderboard page.

source label without registered configuration ID

92.2%

CurrentBenchLM aggregated public evaluation

Source carrier · no additional scoring weight

source-checked

BenchLM public benchmark leaderboards ↗Observed 2026-07-15 · checked 2026-07-15

Artificial Analysis public model evaluation page; exact provider variant as listed on the page.

source label without registered configuration ID

92.1%

CurrentArtificial Analysis evaluation harness

Public reference · not scored

source-checked

Artificial Analysis evaluations for gemini-3-5-flash-medium ↗Observed 2026-07-15 · checked 2026-07-15

Artificial Analysis public model evaluation page; exact provider variant as listed on the page.

source label without registered configuration ID

92.0%

CurrentArtificial Analysis evaluation harness

Public reference · not scored

source-checked

Artificial Analysis evaluations for claude-opus-4-8 ↗Observed 2026-07-15 · checked 2026-07-15

Artificial Analysis public model evaluation page; exact provider variant as listed on the page.

source label without registered configuration ID

92.0%

CurrentArtificial Analysis evaluation harness

Public reference · not scored

source-checked

Artificial Analysis evaluations for gpt-5-4 ↗Observed 2026-07-15 · checked 2026-07-15

Artificial Analysis public model evaluation page; exact provider variant as listed on the page.

source label without registered configuration ID

91.5%

CurrentArtificial Analysis evaluation harness

Public reference · not scored

source-checked

Artificial Analysis evaluations for gpt-5-3-codex ↗Observed 2026-07-15 · checked 2026-07-15

Exact model variant as listed on the BenchLM public benchmark leaderboard page.

source label without registered configuration ID

91.3%

CurrentBenchLM aggregated public evaluation

Source carrier · no additional scoring weight

source-checked

BenchLM public benchmark leaderboards ↗Observed 2026-07-15 · checked 2026-07-15

Exact model variant as listed on the BenchLM public benchmark leaderboard page.

source label without registered configuration ID

91.2%

CurrentBenchLM aggregated public evaluation

Source carrier · no additional scoring weight

source-checked

BenchLM public benchmark leaderboards ↗Observed 2026-07-15 · checked 2026-07-15

Kimi K2.6

source label without registered configuration ID

91.1%

CurrentArtificial Analysis independent evaluation

Public reference · not scored

source-checked

Artificial Analysis Models Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15

Grok 4.20 0309 v2 (Reasoning)

source label without registered configuration ID

91.1%

CurrentArtificial Analysis independent evaluation

Public reference · not scored

source-checked

Artificial Analysis Models Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15

Artificial Analysis public model evaluation page; exact provider variant as listed on the page.

source label without registered configuration ID

91.1%

CurrentArtificial Analysis evaluation harness

Public reference · not scored

source-checked

Artificial Analysis evaluations for gpt-5-6-luna ↗Observed 2026-07-15 · checked 2026-07-15

Artificial Analysis public model evaluation page; exact provider variant as listed on the page.

source label without registered configuration ID

91.1%

CurrentArtificial Analysis evaluation harness

Public reference · not scored

source-checked

Artificial Analysis evaluations for claude-sonnet-5 ↗Observed 2026-07-15 · checked 2026-07-15

Exact model variant as listed on the BenchLM public benchmark leaderboard page.

source label without registered configuration ID

90.4%

CurrentBenchLM aggregated public evaluation

Source carrier · no additional scoring weight

source-checked

BenchLM public benchmark leaderboards ↗Observed 2026-07-15 · checked 2026-07-15

GPT-5.2 (xhigh)

gpt-5-2-xhigh

90.3%

CurrentArtificial Analysis independent evaluation

Public reference · not scored

source-checked

Artificial Analysis Models Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15

Exact model variant as listed on the BenchLM public benchmark leaderboard page.

source label without registered configuration ID

90.3%

CurrentBenchLM aggregated public evaluation

Source carrier · no additional scoring weight

source-checked

BenchLM public benchmark leaderboards ↗Observed 2026-07-15 · checked 2026-07-15

Artificial Analysis public model evaluation page; exact provider variant as listed on the page.

source label without registered configuration ID

90.1%

CurrentArtificial Analysis evaluation harness

Public reference · not scored

source-checked

Artificial Analysis evaluations for grok-4-3 ↗Observed 2026-07-15 · checked 2026-07-15

Exact model variant as listed on the BenchLM public benchmark leaderboard page.

source label without registered configuration ID

90.1%

CurrentBenchLM aggregated public evaluation

Source carrier · no additional scoring weight

source-checked

BenchLM public benchmark leaderboards ↗Observed 2026-07-15 · checked 2026-07-15

Exact model variant as listed on the BenchLM public benchmark leaderboard page.

source label without registered configuration ID

89.9%

CurrentBenchLM aggregated public evaluation

Source carrier · no additional scoring weight

source-checked

BenchLM public benchmark leaderboards ↗Observed 2026-07-15 · checked 2026-07-15

GPT-5.2 Codex (xhigh)

gpt-5-2-codex-xhigh

89.9%

CurrentArtificial Analysis independent evaluation

Public reference · not scored

source-checked

Artificial Analysis Models Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15

GPT-5.4 (high)

gpt-5-4-epoch-gpt-5-4-2026-03-05-high

89.9%

1.0.11Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-07-15 · checked 2026-08-17

Kimi K2.7 Code

source label without registered configuration ID

89.6%

CurrentArtificial Analysis independent evaluation

Public reference · not scored

source-checked

Artificial Analysis Models Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15

Artificial Analysis public model evaluation page; exact provider variant as listed on the page.

source label without registered configuration ID

89.4%

CurrentArtificial Analysis evaluation harness

Public reference · not scored

source-checked

Artificial Analysis evaluations for deepseek-v4-flash ↗Observed 2026-07-15 · checked 2026-07-15

Qwen3.5 397B A17B (Reasoning)

source label without registered configuration ID

89.3%

CurrentArtificial Analysis independent evaluation

Score input

source-checked

Artificial Analysis Models Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15

GPT-5.4 (medium)

gpt-5-4-epoch-gpt-5-4-2026-03-05-medium

88.9%

1.0.11Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-07-15 · checked 2026-08-17

Artificial Analysis public model evaluation page; exact provider variant as listed on the page.

source label without registered configuration ID

88.8%

CurrentArtificial Analysis evaluation harness

Public reference · not scored

source-checked

Artificial Analysis evaluations for deepseek-v4-pro ↗Observed 2026-07-15 · checked 2026-07-15

Qwen3.6 Max Preview

source label without registered configuration ID

Qwen3.6 Max

88.8%

CurrentArtificial Analysis independent evaluation

Public reference · not scored

source-checked

Artificial Analysis Models Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15

Grok 4

source label without registered configuration ID

87.7%

CurrentArtificial Analysis independent evaluation

Public reference · not scored

source-checked

Artificial Analysis Models Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15

Exact model variant as listed on the BenchLM public benchmark leaderboard page.

source label without registered configuration ID

87.6%

CurrentBenchLM aggregated public evaluation

Source carrier · no additional scoring weight

source-checked

BenchLM public benchmark leaderboards ↗Observed 2026-07-15 · checked 2026-07-15

GPT-5.4 mini (xhigh)

gpt-5-4-mini-xhigh

87.5%

CurrentArtificial Analysis independent evaluation

Public reference · not scored

source-checked

Artificial Analysis Models Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15

Exact model variant as listed on the BenchLM public benchmark leaderboard page.

source label without registered configuration ID

87%

CurrentBenchLM aggregated public evaluation

Source carrier · no additional scoring weight

source-checked

BenchLM public benchmark leaderboards ↗Observed 2026-07-15 · checked 2026-07-15

Nemotron 3 Ultra 550B A55B (Reasoning)

source label without registered configuration ID

86.7%

CurrentArtificial Analysis independent evaluation

Score input

source-checked

Artificial Analysis Models Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15

Gemini 3.5 Flash (minimal)

gemini-3-5-flash-epoch-gemini-3-5-flash-minimal

86.4%

1.0.11Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-07-15 · checked 2026-08-17

Qwen3 Max Thinking

source label without registered configuration ID

86.1%

CurrentArtificial Analysis independent evaluation

Public reference · not scored

source-checked

Artificial Analysis Models Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15

Exact model variant as listed on the BenchLM public benchmark leaderboard page.

source label without registered configuration ID

86%

CurrentBenchLM aggregated public evaluation

Source carrier · no additional scoring weight

source-checked

BenchLM public benchmark leaderboards ↗Observed 2026-07-15 · checked 2026-07-15

GLM-4.7 (Reasoning)

source label without registered configuration ID

85.9%

CurrentArtificial Analysis independent evaluation

Public reference · not scored

source-checked

Artificial Analysis Models Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15

Qwen3.5 27B (Reasoning)

source label without registered configuration ID

85.8%

CurrentArtificial Analysis independent evaluation

Public reference · not scored

source-checked

Artificial Analysis Models Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15

Qwen3.5 122B A10B (Reasoning)

source label without registered configuration ID

Qwen3.5 122B

85.7%

CurrentArtificial Analysis independent evaluation

Score input

source-checked

Artificial Analysis Models Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15

GPT-5 (high)

gpt-5-high

85.4%

CurrentArtificial Analysis independent evaluation

Public reference · not scored

source-checked

Artificial Analysis Models Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15

MiMo-V2.5

source label without registered configuration ID

MiMo-V2.5

84.9%

CurrentArtificial Analysis independent evaluation

Public reference · not scored

source-checked

Artificial Analysis Models Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15

MiniMax-M2.5

source label without registered configuration ID

MiniMax M2.5

84.8%

CurrentArtificial Analysis independent evaluation

Public reference · not scored

source-checked

Artificial Analysis Models Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15

GPT-5.4 (low)

gpt-5-4-low

84.8%

1.0.11Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-07-15 · checked 2026-08-17

Grok 4 Fast (Reasoning)

source label without registered configuration ID

Grok 4 Fast

84.7%

CurrentArtificial Analysis independent evaluation

Public reference · not scored

source-checked

Artificial Analysis Models Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15

o3-pro

source label without registered configuration ID

84.5%

CurrentArtificial Analysis independent evaluation

Public reference · not scored

source-checked

Artificial Analysis Models Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15

Qwen3.5 35B A3B (Reasoning)

source label without registered configuration ID

Qwen3.5 35B

84.5%

CurrentArtificial Analysis independent evaluation

Score input

source-checked

Artificial Analysis Models Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15

Gemini 2.5 Pro

source label without registered configuration ID

84.4%

CurrentArtificial Analysis independent evaluation

Public reference · not scored

source-checked

Artificial Analysis Models Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15

Exact model variant as listed on the BenchLM public benchmark leaderboard page.

source label without registered configuration ID

84.3%

CurrentBenchLM aggregated public evaluation

Source carrier · no additional scoring weight

source-checked

BenchLM public benchmark leaderboards ↗Observed 2026-07-15 · checked 2026-07-15

Qwen3.6 27B (Reasoning)

source label without registered configuration ID

84.2%

CurrentArtificial Analysis independent evaluation

Public reference · not scored

source-checked

Artificial Analysis Models Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15

DeepSeek V3.2 (Reasoning)

source label without registered configuration ID

84.0%

CurrentArtificial Analysis independent evaluation

Public reference · not scored

source-checked

Artificial Analysis Models Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15

Kimi K2 Thinking

source label without registered configuration ID

Kimi K2 Thinking

83.8%

CurrentArtificial Analysis independent evaluation

Public reference · not scored

source-checked

Artificial Analysis Models Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15

GPT-5 Codex (high)

gpt-5-codex-high

GPT-5 Codex

83.7%

CurrentArtificial Analysis independent evaluation

Public reference · not scored

source-checked

Artificial Analysis Models Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15

Claude 4.5 Sonnet (Reasoning)

source label without registered configuration ID

83.4%

CurrentArtificial Analysis independent evaluation

Score input

source-checked

Artificial Analysis Models Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15

MiniMax-M2.1

source label without registered configuration ID

MiniMax M2.1

83.0%

CurrentArtificial Analysis independent evaluation

Public reference · not scored

source-checked

Artificial Analysis Models Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15

Exact model variant as listed on the BenchLM public benchmark leaderboard page.

source label without registered configuration ID

82.8%

CurrentBenchLM aggregated public evaluation

Source carrier · no additional scoring weight

source-checked

BenchLM public benchmark leaderboards ↗Observed 2026-07-15 · checked 2026-07-15

source label without registered configuration ID

82.7%

CurrentArtificial Analysis independent evaluation

Public reference · not scored

source-checked

Artificial Analysis Models Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15

Qwen3.5 Omni Plus

source label without registered configuration ID

Qwen3.5 Omni Plus

82.6%

CurrentArtificial Analysis independent evaluation

Public reference · not scored

source-checked

Artificial Analysis Models Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15

Claude 4.1 Opus (Reasoning)

source label without registered configuration ID

80.9%

CurrentArtificial Analysis independent evaluation

Score input

source-checked

Artificial Analysis Models Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15

GPT-5 mini (medium)

gpt-5-mini-medium

80.3%

CurrentArtificial Analysis independent evaluation

Public reference · not scored

source-checked

Artificial Analysis Models Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15

NVIDIA Nemotron 3 Super 120B A12B (Reasoning)

source label without registered configuration ID

Nemotron 3 Super

80%

CurrentArtificial Analysis independent evaluation

Score input

source-checked

Artificial Analysis Models Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15

o3 (low)

o3-epoch-o3-2025-04-16-low

79.8%

1.0.11Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-07-15 · checked 2026-08-17

Claude 4 Opus (Reasoning)

source label without registered configuration ID

79.6%

CurrentArtificial Analysis independent evaluation

Score input

source-checked

Artificial Analysis Models Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15

Gemini 2.5 Flash Preview (Sep '25) (Reasoning)

source label without registered configuration ID

79.3%

CurrentArtificial Analysis independent evaluation

Public reference · not scored

source-checked

Artificial Analysis Models Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15

DeepSeek V3.1 Terminus (Reasoning)

source label without registered configuration ID

DeepSeek V3.1 Terminus

79.2%

CurrentArtificial Analysis independent evaluation

Public reference · not scored

source-checked

Artificial Analysis Models Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15

Gemma 4 26B A4B (Reasoning)

source label without registered configuration ID

Gemma 4 26B

79.2%

CurrentArtificial Analysis independent evaluation

Score input

source-checked

Artificial Analysis Models Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15

Grok 3 mini Reasoning (high)

source label without registered configuration ID

79.1%

CurrentArtificial Analysis independent evaluation

Score input

source-checked

Artificial Analysis Models Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15

Nova 2.0 Pro Preview (medium)

source label without registered configuration ID

Nova 2 Pro

78.5%

CurrentArtificial Analysis independent evaluation

Score input

source-checked

Artificial Analysis Models Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15

o4-mini (high)

o4-mini-high

78.4%

CurrentArtificial Analysis independent evaluation

Public reference · not scored

source-checked

Artificial Analysis Models Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15

GLM-4.6 (Reasoning)

source label without registered configuration ID

78.0%

CurrentArtificial Analysis independent evaluation

Public reference · not scored

source-checked

Artificial Analysis Models Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15

Claude 4 Sonnet (Reasoning)

source label without registered configuration ID

77.7%

CurrentArtificial Analysis independent evaluation

Score input

source-checked

Artificial Analysis Models Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15

MiniMax-M2

source label without registered configuration ID

MiniMax M2

77.7%

CurrentArtificial Analysis independent evaluation

Public reference · not scored

source-checked

Artificial Analysis Models Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15

Claude 3.7 Sonnet (Reasoning)

source label without registered configuration ID

Claude Sonnet 3.7

77.2%

CurrentArtificial Analysis independent evaluation

Score input

source-checked

Artificial Analysis Models Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15

Artificial Analysis public model evaluation page; exact provider variant as listed on the page.

source label without registered configuration ID

76.9%

CurrentArtificial Analysis evaluation harness

Public reference · not scored

source-checked

Artificial Analysis evaluations for mistral-small-4 ↗Observed 2026-07-15 · checked 2026-07-15

Kimi K2 0905

source label without registered configuration ID

Kimi K2 0905

76.7%

CurrentArtificial Analysis independent evaluation

Public reference · not scored

source-checked

Artificial Analysis Models Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15

Command A+

source label without registered configuration ID

76.1%

CurrentArtificial Analysis independent evaluation

Public reference · not scored

source-checked

Artificial Analysis Models Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15

Gemma 4 12B (Reasoning)

source label without registered configuration ID

75.3%

CurrentArtificial Analysis independent evaluation

Public reference · not scored

source-checked

Artificial Analysis Models Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15

Ling-2.6-1T

source label without registered configuration ID

Ling 2.6 1T

75.2%

CurrentArtificial Analysis independent evaluation

Public reference · not scored

source-checked

Artificial Analysis Models Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15

Artificial Analysis public model evaluation page; exact provider variant as listed on the page.

source label without registered configuration ID

Mistral Medium 3.5

74.8%

CurrentArtificial Analysis evaluation harness

Public reference · not scored

source-checked

Artificial Analysis evaluations for mistral-medium-3-5 ↗Observed 2026-07-15 · checked 2026-07-15

source label without registered configuration ID

74.7%

CurrentArtificial Analysis independent evaluation

Public reference · not scored

source-checked

Artificial Analysis Models Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15

GPT-5.4 (none)

gpt-5-4-epoch-gpt-5-4-2026-03-05-none

74.7%

1.0.11Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-07-15 · checked 2026-08-17

Exact model variant as listed on the BenchLM public benchmark leaderboard page.

source label without registered configuration ID

72.9%

CurrentBenchLM aggregated public evaluation

Source carrier · no additional scoring weight

source-checked

BenchLM public benchmark leaderboards ↗Observed 2026-07-15 · checked 2026-07-15

Grok Code Fast 1

source label without registered configuration ID

72.7%

CurrentArtificial Analysis independent evaluation

Public reference · not scored

source-checked

Artificial Analysis Models Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15

Exact model variant as listed on the BenchLM public benchmark leaderboard page.

source label without registered configuration ID

71.2%

CurrentBenchLM aggregated public evaluation

Source carrier · no additional scoring weight

source-checked

BenchLM public benchmark leaderboards ↗Observed 2026-07-15 · checked 2026-07-15

Artificial Analysis public model evaluation page; exact provider variant as listed on the page.

source label without registered configuration ID

68.0%

CurrentArtificial Analysis evaluation harness

Public reference · not scored

source-checked

Artificial Analysis evaluations for mistral-large-3 ↗Observed 2026-07-15 · checked 2026-07-15

Artificial Analysis public model evaluation page; exact provider variant as listed on the page.

source label without registered configuration ID

64.7%

CurrentArtificial Analysis evaluation harness

Score input

source-checked

Artificial Analysis evaluations for claude-4-5-haiku ↗Observed 2026-07-15 · checked 2026-07-15

grok-4.20-0309-reasoning

grok-4-20-epoch-grok-4-20-0309-reasoning

89.3%

1.0.11Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-07-13 · checked 2026-08-17

Claude Sonnet 4.6 (high)

claude-sonnet-4-6-non-reasoning-high

83.3%

1.0.11Epoch AI benchmark runner

Score input

source-checked

Epoch AI permanent refresh source ↗Observed 2026-07-13 · checked 2026-08-17

Claude Sonnet 4.6 (medium)

claude-sonnet-4-6-livebench-2026-06-25-medium

83.3%

1.0.11Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-07-13 · checked 2026-08-17

GPT-5.2 (none)

gpt-5-2-epoch-gpt-5-2-2025-12-11-none

73.2%

1.0.11Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-07-13 · checked 2026-08-17

GPT-5 nano (low)

gpt-5-nano-epoch-gpt-5-nano-2025-08-07-low

57.6%

1.0.11Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-07-13 · checked 2026-08-18

GPT-5 nano (low)

gpt-5-nano-epoch-gpt-5-nano-2025-08-07-minimal

48.5%

1.0.11Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-07-13 · checked 2026-08-18

o4-mini (low)

o4-mini-epoch-o4-mini-2025-04-16-low

75.3%

1.0.11Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-07-11 · checked 2026-08-17

GPT-5.6 Sol (max)

gpt-5-6-sol-max

93.5%

1.0.11Epoch AI benchmark runner

Score input

source-checked

Epoch AI permanent refresh source ↗Observed 2026-07-09 · checked 2026-08-17

GPT-5.6 Terra (max)

gpt-5-6-terra-max

93.3%

1.0.11Epoch AI benchmark runner

Score input

source-checked

Epoch AI permanent refresh source ↗Observed 2026-07-09 · checked 2026-08-17

GPT-5.6 Luna (max)

gpt-5-6-luna-max

91.6%

1.0.11Epoch AI benchmark runner

Score input

source-checked

Epoch AI permanent refresh source ↗Observed 2026-07-09 · checked 2026-08-17

Grok 4.5 (high)

grok-4-5-aa-2-high

93.4%

1.0.11Epoch AI benchmark runner

Score input

source-checked

Epoch AI permanent refresh source ↗Observed 2026-07-08 · checked 2026-08-17

tencent/Hy3 instruct weights; configuration as recorded on the Hugging Face evaluation results entry.

source label without registered configuration ID

90.4%

CurrentProvider / community evaluation result attached on Hugging Face

Score input

provider-reported

Tencent Hy3 model card on Hugging Face ↗Observed 2026-07-06 · checked 2026-07-16

Claude Sonnet 5 (xhigh)

claude-sonnet-5-xhigh

90.5%

1.0.11Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-07-01 · checked 2026-08-17

LongCat-2.0 (in-house harness)

longcat-2-0-default

LongCat-2.0

88.9%

DiamondLongCat unified in-house harness

Score input

provider-reported

LongCat-2.0 model card ↗Observed 2026-06-30 · checked 2026-08-15

GLM-5.2 (max)

glm-5-2-max

91.9%

1.0.11Epoch AI benchmark runner

Score input

source-checked

Epoch AI permanent refresh source ↗Observed 2026-06-24 · checked 2026-08-17

grok-4.3_high

grok-4-3-epoch-grok-4-3-high

88.8%

1.0.11Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-06-17 · checked 2026-08-17

DeepSeek v4 (max)

deepseek-v4-pro-max

89.6%

1.0.11Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-06-16 · checked 2026-08-18

Claude Opus 4.8 (max)

claude-opus-4-8-max

91.0%

1.0.10Epoch AI benchmark runner

Score input

source-checked

Epoch AI permanent refresh source ↗Observed 2026-06-07 · checked 2026-08-17

Gemini 3.5 Flash (high)

gemini-3-5-flash-high

92.8%

1.0.9Epoch AI benchmark runner

Score input

source-checked

Epoch AI permanent refresh source ↗Observed 2026-05-22 · checked 2026-08-17

GPT-5.5 (low)

gpt-5-5-low

90.7%

1.0.8Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-05-05 · checked 2026-08-17

Kimi K2.6

kimi-k2-6-epoch-kimi-k2-6

90.8%

1.0.8Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-05-01 · checked 2026-08-17

GPT-5.5 (xhigh)

gpt-5-5-xhigh

94.0%

1.0.8Epoch AI benchmark runner

Score input

source-checked

Epoch AI permanent refresh source ↗Observed 2026-04-24 · checked 2026-08-17

GPT-5.5 Pro (xhigh)

gpt-5-5-pro

93.9%

1.0.8Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-04-24 · checked 2026-08-17

Claude Opus 4.7 (xhigh)

claude-opus-4-7-livebench-2026-06-25-xhigh

90.2%

1.0.7Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-04-17 · checked 2026-08-17

GPT-5.4 mini (high)

gpt-5-4-mini-epoch-gpt-5-4-mini-2026-03-17-high

83.6%

1.0.6Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-04-15 · checked 2026-08-17

GPT-5.4 nano (high)

gpt-5-4-nano-epoch-gpt-5-4-nano-2026-03-17-high

78.5%

1.0.6Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-04-14 · checked 2026-08-17

GPT-5.4 Pro (xhigh)

gpt-5-4-pro-xhigh

94.6%

1.0.6Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-03-20 · checked 2026-08-18

GPT-5.4 (xhigh)

gpt-5-4-xhigh

93.3%

1.0.6Epoch AI benchmark runner

Score input

source-checked

Epoch AI permanent refresh source ↗Observed 2026-03-06 · checked 2026-08-17

Gemini 3.1 Pro Preview

gemini-3-1-pro-preview-epoch-gemini-3-1-pro-preview

94.1%

1.0.5Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-02-20 · checked 2026-08-17

Claude Sonnet 4.6 (32k thinking)

claude-sonnet-4-6-epoch-claude-sonnet-4-6-32k

87.4%

1.0.6Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-02-20 · checked 2026-08-17

GLM-5

glm-5-epoch-glm-5

87.8%

1.0.5Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-02-12 · checked 2026-08-17

Claude Opus 4.6 (32k thinking)

claude-opus-4-6-epoch-claude-opus-4-6-32k

90.5%

1.0.4Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-02-06 · checked 2026-08-17

Claude Opus 4.6 (64k thinking)

claude-opus-4-6-epoch-claude-opus-4-6-64k

88.8%

1.0.4Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-02-06 · checked 2026-08-17

Kimi K2.5 (Fireworks)

kimi-k2-5-epoch-fireworks-kimi-k2p5

87.6%

1.0.4Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-02-02 · checked 2026-08-17

GLM-4.7

glm-4-7-epoch-glm-4-7

83.3%

1.0.4Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2026-01-29 · checked 2026-08-17

Kimi K2.5 thinking configuration

kimi-k2-5-thinking

87.6%

diamondKimi K2.5 official model-card evaluation

Source carrier · no additional scoring weight

source-checked

Kimi K2.5 official model card and evaluation table ↗Observed 2026-01-27 · checked 2026-08-29

Gemini 3 Flash Preview

gemini-3-flash-epoch-gemini-3-flash-preview

83.2%

1.0.3Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-12-17 · checked 2026-08-17

DeepSeek-V3.2 (Thinking)

deepseek-v3-2-epoch-deepseek-reasoner

83.4%

1.0.3Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-12-16 · checked 2026-08-17

GPT-5.2 (xhigh)

gpt-5-2-xhigh

91.4%

1.0.3Epoch AI benchmark runner

Score input

source-checked

Epoch AI permanent refresh source ↗Observed 2025-12-13 · checked 2026-08-17

GPT-5.2 (high)

gpt-5-2-livebench-2026-06-25-high

88.2%

1.0.3Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-12-11 · checked 2026-08-17

GPT-5.2 (medium)

gpt-5-2-medium

87.9%

1.0.3Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-12-11 · checked 2026-08-17

GPT-5.2 (low)

gpt-5-2-epoch-gpt-5-2-2025-12-11-low

82.7%

1.0.3Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-12-11 · checked 2026-08-17

gpt-oss-120b (high)

gpt-oss-120b-high

75.8%

1.0.3Epoch AI benchmark runner

Score input

source-checked

Epoch AI permanent refresh source ↗Observed 2025-12-11 · checked 2026-08-18

Claude Opus 4.5 (16k thinking)

claude-opus-4-5-epoch-claude-opus-4-5-20251101-16k

85.5%

1.0.2Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-11-25 · checked 2026-08-17

Claude Opus 4.5 (32k thinking)

claude-opus-4-5-epoch-claude-opus-4-5-20251101-32k

86.0%

1.0.2Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-11-24 · checked 2026-08-17

Claude Opus 4.5 (no thinking)

claude-opus-4-5-epoch-claude-opus-4-5-20251101

80.7%

1.0.2Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-11-24 · checked 2026-08-17

Gemini 3 Pro Preview

gemini-3-pro-epoch-gemini-3-pro-preview

Gemini 3 Pro Preview

92.6%

1.0.2Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-11-19 · checked 2026-08-18

GPT-5.1 (medium)

gpt-5-1-epoch-gpt-5-1-2025-11-13-medium

85.0%

1.0.1Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-11-17 · checked 2026-08-17

Gemini 2.5 Pro

gemini-2-5-pro-epoch-gemini-2-5-pro

85.3%

1.0.1Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-11-16 · checked 2026-08-17

GPT-5.1 (high)

gpt-5-1-high

87.6%

1.0.1Epoch AI benchmark runner

Score input

source-checked

Epoch AI permanent refresh source ↗Observed 2025-11-13 · checked 2026-08-17

Kimi K2 Thinking Turbo

kimi-k2-thinking-epoch-kimi-k2-thinking-turbo

Kimi K2 Thinking

84.2%

1.0.1Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-11-11 · checked 2026-08-17

GPT-5 mini (high)

gpt-5-mini-high

75%

1.0.1Epoch AI benchmark runner

Score input

source-checked

Epoch AI permanent refresh source ↗Observed 2025-10-30 · checked 2026-08-17

GPT-5 nano (high)

gpt-5-nano-high

69.4%

1.0.1Epoch AI benchmark runner

Score input

source-checked

Epoch AI permanent refresh source ↗Observed 2025-10-30 · checked 2026-08-18

GPT-5 (high)

gpt-5-high

86.2%

1.0.1Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-10-29 · checked 2026-08-17

Claude Sonnet 4.5 (59k thinking)

claude-sonnet-4-5-epoch-claude-sonnet-4-5-20250929-59k

82.3%

1.0.0Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-10-28 · checked 2026-08-17

Claude Sonnet 4.5 (16k thinking)

claude-sonnet-4-5-epoch-claude-sonnet-4-5-20250929-16k

78.8%

1.0.0Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-10-28 · checked 2026-08-17

Claude Haiku 4.5 (32k thinking)

claude-haiku-4-5-epoch-claude-haiku-4-5-20251001-32k

71.2%

1.0.0Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-10-22 · checked 2026-08-17

Claude Sonnet 4.5 (32k thinking)

claude-sonnet-4-5-epoch-claude-sonnet-4-5-20250929-32k

81.7%

1.0.0Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-10-21 · checked 2026-08-17

claude-haiku-4-5-20251001

claude-haiku-4-5-epoch-claude-haiku-4-5-20251001

60.5%

1.0.0Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-10-16 · checked 2026-08-17

Qwen3 Max

qwen3-max-epoch-qwen3-max-2025-09-23

72.6%

1.0.0Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-10-06 · checked 2026-08-17

Claude Sonnet 4.5 (no thinking)

claude-sonnet-4-5-epoch-claude-sonnet-4-5-20250929

73.7%

1.0.0Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-09-29 · checked 2026-08-17

GPT-5 (medium)

gpt-5-medium

85.4%

1.0.0Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-08-07 · checked 2026-08-17

GPT-5 mini (medium)

gpt-5-mini-medium

71.7%

1.0.0Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-08-07 · checked 2026-08-17

GPT-5 nano (medium)

gpt-5-nano-medium

67.4%

1.0.0Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-08-07 · checked 2026-08-18

Claude Opus 4.1 (16k thinking)

claude-opus-4-1-epoch-claude-opus-4-1-20250805-16k

77.3%

1.0.0Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-08-05 · checked 2026-08-18

Claude Opus 4.1 (27k thinking)

claude-opus-4-1-epoch-claude-opus-4-1-20250805-27k

76.8%

1.0.0Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-08-05 · checked 2026-08-18

Claude Opus 4.1

claude-opus-4-1-epoch-claude-opus-4-1-20250805

73.2%

1.0.0Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-08-05 · checked 2026-08-18

Gemini 2.5 Pro Preview (Jun 2025)

gemini-2-5-pro-epoch-gemini-2-5-pro-preview-06-05

84.8%

1.0.0Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-06-05 · checked 2026-08-17

Claude Sonnet 4 (59k thinking)

claude-sonnet-4-epoch-claude-sonnet-4-20250514-59k

79.2%

1.0.0Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-05-26 · checked 2026-08-18

Grok 3 (beta)

grok-3-epoch-grok-3-beta

Grok 3

75.8%

1.0.0Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-05-26 · checked 2026-08-18

grok-3-mini-beta_high

grok-3-mini-epoch-grok-3-mini-beta-high

75.5%

1.0.0Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-05-26 · checked 2026-08-17

DeepSeek-R1

deepseek-r1-epoch-deepseek-r1

71.7%

1.0.0Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-05-26 · checked 2026-08-18

Claude Sonnet 4 (32k thinking)

claude-sonnet-4-epoch-claude-sonnet-4-20250514-32k

78.3%

1.0.0Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-05-22 · checked 2026-08-18

Claude Opus 4 (16k thinking)

claude-opus-4-epoch-claude-opus-4-20250514-16k

76.3%

1.0.0Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-05-22 · checked 2026-08-18

Claude Sonnet 4 (16k thinking)

claude-sonnet-4-epoch-claude-sonnet-4-20250514-16k

75.8%

1.0.0Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-05-22 · checked 2026-08-18

Claude Opus 4

claude-opus-4-epoch-claude-opus-4-20250514

69.2%

1.0.0Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-05-22 · checked 2026-08-18

Claude Sonnet 4

claude-sonnet-4-epoch-claude-sonnet-4-20250514

66.7%

1.0.0Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-05-22 · checked 2026-08-18

mistral-medium-2505

mistral-medium-3-epoch-mistral-medium-2505

59.5%

1.0.0Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-05-07 · checked 2026-08-18

o3 (high)

o3-epoch-o3-2025-04-16-high

81.8%

1.0.0Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-04-16 · checked 2026-08-17

o4-mini (high)

o4-mini-high

79.6%

1.0.0Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-04-16 · checked 2026-08-17

GPT-4.1

gpt-4-1-epoch-gpt-4-1-2025-04-14

66.9%

1.0.0Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-04-14 · checked 2026-08-18

GPT-4.1 mini

gpt-4-1-mini-epoch-gpt-4-1-mini-2025-04-14

65.8%

1.0.0Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-04-14 · checked 2026-08-18

GPT-4.1 nano

gpt-4-1-nano-epoch-gpt-4-1-nano-2025-04-14

48.9%

1.0.0Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-04-14 · checked 2026-08-18

grok-3-mini-beta_low

grok-3-mini-epoch-grok-3-mini-beta-low

76.3%

1.0.0Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-04-10 · checked 2026-08-17

Llama 4 Maverick (FP8)

llama-4-maverick-epoch-llama-4-maverick-17b-128e-instruct-fp8

67.0%

1.0.0Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-04-08 · checked 2026-08-18

Llama-4-Scout-17B-16E-Instruct

llama-4-scout-epoch-llama-4-scout-17b-16e-instruct

DeepSeek R1 Distill Llama 70B

51.8%

1.0.0Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-04-08 · checked 2026-08-18

Qwen Turbo

qwen-turbo-epoch-qwen-turbo-2024-11-01

Qwen2.5 Turbo

41.8%

1.0.0Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-04-07 · checked 2026-08-18

Qwen2.5-Max

qwen-2-5-max-epoch-qwen-max-2025-01-25

Qwen2.5 Max

56.1%

1.0.0Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-04-01 · checked 2026-08-18

mistral-small-2503

mistral-small-3-1-epoch-mistral-small-2503

Mistral Small 3.1

47.5%

1.0.0Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-03-18 · checked 2026-08-18

Claude 3.5 Haiku (Oct 2024)

claude-3-5-haiku-epoch-claude-3-5-haiku-20241022

Claude 3.5 Haiku

38.1%

1.0.0Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-03-12 · checked 2026-08-18

gemini-1.5-flash-8b-001

gemini-1-5-flash-8b-epoch-gemini-1-5-flash-8b-001

Gemini 1.5 Flash-8B

33.0%

1.0.0Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-03-12 · checked 2026-08-18

DeepSeek-R1-Distill-Llama-70B

deepseek-r1-distill-llama-70b-epoch-deepseek-r1-distill-llama-70b

55.7%

1.0.0Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-03-10 · checked 2026-08-18

DeepSeek-R1-Distill-Qwen-14B

deepseek-r1-distill-qwen-14b-epoch-deepseek-r1-distill-qwen-14b

DeepSeek R1 Distill Qwen 14B

44.7%

1.0.0Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-03-10 · checked 2026-08-18

GPT-4.5 Preview (Feb 2025)

gpt-4-5-epoch-gpt-4-5-preview-2025-02-27

GPT-4.5 (Preview)

68.7%

1.0.0Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-02-28 · checked 2026-08-18

mistral-large-2411

mistral-large-2-epoch-mistral-large-2411

51.3%

1.0.0Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-02-25 · checked 2026-08-18

o3-mini (high)

o3-mini-high

Gemini 2.0 Flash (Feb '25)

77.0%

1.0.0Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-02-13 · checked 2026-08-18

o1 (high)

o1-epoch-o1-2024-12-17-high

76.8%

1.0.0Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-02-13 · checked 2026-08-17

o1-mini (high)

o1-mini-epoch-o1-mini-2024-09-12-high

o1-mini

62.4%

1.0.0Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-02-13 · checked 2026-08-18

Gemini 2.0 Flash (Feb 2025)

gemini-2-0-flash-epoch-gemini-2-0-flash-001

64.1%

1.0.0Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-02-06 · checked 2026-08-18

GPT-4o (Nov 2024)

gpt-4o-epoch-gpt-4o-2024-11-20

47.9%

1.0.0Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-02-05 · checked 2026-08-18

o3-mini (medium)

o3-mini-epoch-o3-mini-2025-01-31-medium

74.3%

1.0.0Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-01-31 · checked 2026-08-18

phi-4

phi-4-epoch-phi-4

56.1%

1.0.0Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-01-31 · checked 2026-08-18

mistral-small-2501

mistral-small-3-epoch-mistral-small-2501

Mistral Small 3

45.3%

1.0.0Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-01-30 · checked 2026-08-18

o1 (medium)

o1-epoch-o1-2024-12-17-medium

75.8%

1.0.0Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-01-27 · checked 2026-08-17

o1-mini (medium)

o1-mini-epoch-o1-mini-2024-09-12-medium

o1-mini

59.5%

1.0.0Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-01-27 · checked 2026-08-18

gemini-1.5-pro-002

gemini-1-5-pro-epoch-gemini-1-5-pro-002

57.2%

1.0.0Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-01-27 · checked 2026-08-18

DeepSeek-V3

deepseek-v3-epoch-deepseek-v3

56.5%

1.0.0Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-01-27 · checked 2026-08-18

Claude 3.5 Sonnet (Jun 2024)

claude-3-5-sonnet-epoch-claude-3-5-sonnet-20240620

Claude 3.5 Sonnet

54.0%

1.0.0Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-01-27 · checked 2026-08-18

Llama-3.1-405B-Instruct

llama-3-1-405b-epoch-llama-3-1-405b-instruct

50.9%

1.0.0Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-01-27 · checked 2026-08-18

o1-preview

o1-preview-epoch-o1-preview-2024-09-12

o1-preview

50.3%

1.0.0Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-01-27 · checked 2026-08-18

GPT-4o (Aug 2024)

gpt-4o-epoch-gpt-4o-2024-08-06

49.2%

1.0.0Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-01-27 · checked 2026-08-18

Qwen2.5-72B

qwen2-5-72b-epoch-qwen2-5-72b-instruct

Qwen2.5-72B

49.1%

1.0.0Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-01-27 · checked 2026-08-18

mistral-large-2407

mistral-large-2-epoch-mistral-large-2407

49.0%

1.0.0Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-01-27 · checked 2026-08-18

GPT-4o (May 2024)

gpt-4o-epoch-gpt-4o-2024-05-13

Gemini 1.5 Flash (Sep '24)

48.9%

1.0.0Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-01-27 · checked 2026-08-18

gemini-1.5-flash-002

gemini-1-5-flash-epoch-gemini-1-5-flash-002

47.3%

1.0.0Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-01-27 · checked 2026-08-18

Claude 3 Opus

claude-3-opus-epoch-claude-3-opus-20240229

47.2%

1.0.0Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-01-27 · checked 2026-08-18

gemini-1.5-pro-001

gemini-1-5-pro-epoch-gemini-1-5-pro-001

Gemini 1.5 Flash (Sep '24)

45.9%

1.0.0Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-01-27 · checked 2026-08-18

Claude 3 Sonnet

claude-3-sonnet-epoch-claude-3-sonnet-20240229

Claude 3 Sonnet

40.6%

1.0.0Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-01-27 · checked 2026-08-18

Meta-Llama-3-70B-Instruct

llama-3-70b-epoch-meta-llama-3-70b-instruct

Llama 3 70B

40.6%

1.0.0Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-01-27 · checked 2026-08-18

Gemini 1.5 Flash (May 2024)

gemini-1-5-flash-epoch-gemini-1-5-flash-001

40.4%

1.0.0Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-01-27 · checked 2026-08-18

mistral-large-2402

mistral-large-epoch-mistral-large-2402

Mistral Large (Feb '24)

38.8%

1.0.0Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-01-27 · checked 2026-08-18

gpt-4o-mini-2024-07-18

gpt-4o-mini-epoch-gpt-4o-mini-2024-07-18

37.7%

1.0.0Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-01-27 · checked 2026-08-18

Claude 3 Haiku

claude-3-haiku-epoch-claude-3-haiku-20240307

36.3%

1.0.0Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-01-27 · checked 2026-08-18

claude-2.0

claude-2-epoch-claude-2-0

Claude 2.0

34.7%

1.0.0Epoch AI benchmark runner

Public reference · not scored

source-checked

Epoch AI permanent refresh source ↗Observed 2025-01-27 · checked 2026-08-18

gemini-1.0-pro-001

gemini-1-0-pro-epoch-gemini-1-0-pro-001