MMMU-Pro

gemini-3.5-flash

Google

closed

Source carrier

84.3%

gemini-3.1-pro-preview

Google

closed

Source carrier

83.9%

gpt-5.6-sol

OpenAI

closed

Score input

83.4%

gemini-3.6-flash

Google

closed

Score input

83.2%

Qwen3.8 Max

qwen3.8-max

Alibaba Cloud

open

Score input

82.3%

kimi-k3

Moonshot AI

open

Score input

81.6%

gpt-5.4

OpenAI

closed

Source carrier

81.2%

gpt-5.5

OpenAI

closed

Public reference

81.2%

#10

gpt-5.6-terra

OpenAI

closed

Source carrier

80.7%

#11

muse-spark

Meta

closed

Source carrier

80.5%

#12

qwen3.7-plus

Alibaba Cloud

closed

Source carrier

80.5%

#13

grok-4.5

xAI

closed

Score input

80.4%

#14

kimi-k2.6

Moonshot AI

open

Source carrier

79.4%

#15

gemini-3.5-flash-lite

Google

closed

Score input

79%

#16

claude-opus-4-7

Anthropic

closed

Source carrier

78.8%

#17

qwen3-6-plus

Alibaba Cloud

closed

Source carrier

78.8%

#18

gemini-3-flash

Google

closed

Source carrier

78.6%

#19

gpt-5.6-luna

OpenAI

closed

Source carrier

78.6%

#20

MiniMax-M3

MiniMax

open

Source carrier

78.6%

#21

gpt-5.3-codex

OpenAI

closed

Score input

78.5%

#22

kimi-k2.5

Moonshot AI

open

Source carrier

78.5%

#23

grok-4.3

xAI

closed

Source carrier

78.1%

#24

claude-opus-4-6

Anthropic

closed

Source carrier

77.3%

#25

claude-sonnet-5

Anthropic

closed

Source carrier

77.3%

#26

Qwen/Qwen3.5-397B-A17B

Alibaba Cloud

open

Source carrier

77.3%

#27

gemma-4-31b

Google

open

Source carrier

76.9%

#28

gpt-5.2-codex

OpenAI

closed

Score input

76.3%

#29

gemini-3.1-flash-lite

Google

closed

Source carrier

75.5%

#30

gpt-5.1

OpenAI

closed

Source carrier

75.5%

#31

MiMo-V2.5

mimo-v2.5

Xiaomi

open

Score input

75.4%

#32

step-3-7-flash

StepFun

open

Source carrier

75.3%

#33

qwen3.5-27b

Alibaba Cloud

open

Score input

75.0%

#34

Qwen/Qwen3.5-122B-A10B

Alibaba Cloud

open

Source carrier

75%

#35

Qwen/Qwen3.6-35B-A3B

Alibaba Cloud

open

Source carrier

75%

#36

Qwen3.5 122B

qwen3.5-122b-a10b

Alibaba Cloud

open

Score input

75.0%

#37

gemini-2.5-pro

Google

closed

Score input

74.9%

#38

qwen3.6-27b

Alibaba Cloud

open

Score input

74.6%

#39

Grok 4.20

grok-4.20

xAI

closed

Score input

74.6%

#40

gpt-5

OpenAI

closed

Source carrier

74.3%

#41

claude-opus-4-5-thinking

Anthropic

closed

Source carrier

74%

#42

Muse Glimmer 30B

meta-models/Muse-Glimmer-30B

Meta

open

Score input

74%

#43

GPT-5 Codex

gpt-5-codex

OpenAI

closed

Score input

73.8%

#44

inkling

Thinking Machines Lab

open

Source carrier

73.5%

#45

gpt-5.4-mini

OpenAI

closed

Source carrier

73.3%

#46

gemini-2.5-flash

Google

closed

Score input

73.1%

#47

glm-5v-turbo

Z.AI

closed

Source carrier

72.8%

#48

qwen3-5-35b-a3b

Alibaba Cloud

open

Source carrier

72.7%

#49

Qwen3.5 35B

qwen3.5-35b-a3b

Alibaba Cloud

open

Score input

72.7%

#50

gpt-5-1-codex

OpenAI

closed

Source carrier

72.5%

#51

gpt-5-1-codex-max

OpenAI

closed

Source carrier

72.5%

#52

claude-opus-4-5

Anthropic

closed

Source carrier

71.2%

#53

claude-sonnet-4-6

Anthropic

closed

Source carrier

70.6%

#54

Qwen3.5 Omni Plus

qwen3.5-omni-plus

Alibaba Cloud

closed

Score input

70.5%

#55

OpenAI

closed

Source carrier

70.1%

#56

mimo-v2-omni

Xiaomi

closed

Source carrier

69.9%

#57

google/gemma-4-12B

Google

open

Source carrier

69.7%

#58

Gemma 4 26B

gemma-4-26b-a4b

Google

open

Score input

69.2%

#59

o4-mini

OpenAI

closed

Score input

69.2%

#60

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

Google

open

Source carrier

69.2%

#61

GPT-5 mini

gpt-5-mini

OpenAI

closed

Score input

68.8%

#62

grok-4

xAI

closed

Score input

68.8%

#63

Claude Sonnet 4.5

claude-sonnet-4-5

Anthropic

closed

Score input

68.7%

#64

Claude Opus 4.1

claude-opus-4-1

Anthropic

closed

Score input

67.9%

#65

claude-4-1-opus-thinking

Anthropic

closed

Source carrier

67.9%

#66

gpt-5.4-nano

OpenAI

closed

Source carrier

66.1%

#67

mistralai/Mistral-Medium-3.5-128B

Mistral AI

open

Source carrier

64.9%

#68

Mistral Medium 3.5

mistral-medium-3-5

Mistral AI

open

Score input

64.9%

#69

Nova 2 Pro

nova-2-pro-preview

Amazon

closed

Score input

64.5%

#70

grok-4-1-fast-reasoning

xAI

closed

Source carrier

63.3%

#71

command-a-plus

Cohere

open

Score input

63.2%

#72

claude-4-sonnet

Anthropic

closed

Source carrier

62.4%

#73

llama-4-maverick

Meta

open

Source carrier

62.1%

#74

grok-4-fast-reasoning

xAI

closed

Source carrier

61.8%

#75

Claude Sonnet 4

claude-sonnet-4

Anthropic

closed

Score input

61.8%

#76

Grok 4 Fast

grok-4-fast

xAI

closed

Score input

61.8%

#77

gpt-4-1

OpenAI

closed

Source carrier

61.2%

#78

gpt-4-1-mini

OpenAI

closed

Source carrier

58.7%

#79

mistral-small-2603

Mistral AI

open

Score input

56.8%

#80

mistral-large-2512

Mistral AI

open

Source carrier

55.7%

#81

Claude Haiku 4.5

claude-haiku-4-5-20251001

Anthropic

closed

Score input

55.1%

#82

gemini-1-5-pro

Google

closed

Source carrier

55%

#83

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

NVIDIA

open

Source carrier

53.2%

#84

mistral-medium-3

Mistral AI

closed

Source carrier

53%

#85

llama-4-scout

Meta

open

Source carrier

52.9%

#86

GPT-4o (May '24)

gpt-4o-2024-05-13

OpenAI

closed

Score input

51.9%

#87

gemma-4-e4b

Google

open

Source carrier

51.4%

#88

grok-4-1-fast

xAI

closed

Source carrier

48.4%

#89

gemma-3-27b

Google

open

Source carrier

48%

#90

gemma-4-e2b

Google

open

Source carrier

44.6%

#91

nova-pro

Amazon

closed

Source carrier

44.3%

#92

Gemini 1.5 Pro (May '24)

gemini-1-5-pro-may-2024

Google

—

Score input

43.5%

#93

gpt-4o-mini

OpenAI

closed

Source carrier

41.5%

#94

gpt-4-1-nano

OpenAI

closed

Source carrier

40.1%

#95

claude-3-haiku

Anthropic

closed

Source carrier

30.8%

#96

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

LiquidAI

open

Source carrier

26.5%

Every distinct dated source-published row for MMMU-Pro, including exact model configuration labels
Exact published model / configuration	Canonical product	Score	Version / system	Evidence	Source / observed
Muse Glimmer-30B; high reasoning; temperature=1.0; top_p=0.95; top_k=64 muse-glimmer-30b-high	Muse Glimmer 30B	74%	1,730-question setArtificial Analysis MMMU-Pro documented evaluation	Public reference · not scored provider-reported	Muse Glimmer Evaluation Methodology ↗Observed 2026-08-10 · checked 2026-08-10
Qwen3.8 Max (xhigh, default reasoning effort) qwen-3-8-max-xhigh	Qwen3.8 Max	82.3%	CurrentQwen provider evaluation	Public reference · not scored provider-reported	Qwen3.8 Max official release ↗Observed 2026-08-03 · checked 2026-08-05
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	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 Gemini 3.5 Flash; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Gemini 3.5 Flash	84.3%	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-5.6 Sol; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.6 Sol	83.4%	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 Gemini 3.6 Flash; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Gemini 3.6 Flash	83.2%	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-5.6 Terra; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.6 Terra	80.7%	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 K3; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Kimi K3	80.5%	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 Muse Spark; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Muse Spark	80.5%	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 Qwen3.7 Plus; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Qwen3.7-Plus	80.5%	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 4.5; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Grok 4.5	80.4%	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-5.5; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.5	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 Kimi K2.6; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Kimi K2.6	79.4%	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 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	79%	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 (Adaptive); bulk export does not retain a complete upstream harness configuration. claude-opus-4-7-max	Claude Opus 4.7	78.8%	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 Gemini 3 Flash; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Gemini 3 Flash	78.6%	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-5.6 Luna; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.6 Luna	78.6%	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 MiniMax M3; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	MiniMax M3	78.6%	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-5.3 Codex; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.3-Codex	78.5%	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-5.4; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.4	78.4%	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 4.3; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Grok 4.3	78.1%	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 Qwen3.6 Plus; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Qwen3.6 Plus	78%	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 5; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude Sonnet 5	77.3%	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; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Qwen3.5 397B A17B	77.3%	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	77.3%	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 4.7; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude Opus 4.7	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 GPT-5.2-Codex; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.2 Codex	76.3%	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 3.1 Flash-Lite; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Gemini 3.1 Flash-Lite	75.5%	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-5.1; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.1	75.5%	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 Opus 4.6 (Adaptive); bulk export does not retain a complete upstream harness configuration. claude-opus-4-6-max	Claude Opus 4.6	75.4%	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 Kimi K2.5; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Kimi K2.5	75.4%	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 Kimi K2.5 (Reasoning); bulk export does not retain a complete upstream harness configuration. kimi-k2-5-thinking	Kimi K2.5	75.4%	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 Step 3.7 Flash; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Step 3.7 Flash	75.3%	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 Qwen3.5-122B-A10B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Qwen3.5-122B-A10B	75%	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	75%	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	75%	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	74.9%	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 Qwen3.6-27B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Qwen3.6 27B	74.6%	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-5 (medium); bulk export does not retain a complete upstream harness configuration. gpt-5-medium	GPT-5	74.3%	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-5 (high); bulk export does not retain a complete upstream harness configuration. gpt-5-high	GPT-5	74.2%	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 Thinking; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude Opus 4.5 Thinking	74%	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; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Inkling	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 Gemma 4 31B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Gemma 4 31B	73.4%	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-5.4 mini; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.4 mini	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 GLM-5V-Turbo; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GLM-5V-Turbo	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 Qwen3.5-35B-A3B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Qwen3.5-35B-A3B	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 Claude Opus 4.6; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude Opus 4.6	72.5%	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-5.1-Codex; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.1-Codex	72.5%	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-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	72.5%	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 Opus 4.5; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude Opus 4.5	71.2%	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 Sonnet 4.6; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude Sonnet 4.6	70.6%	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 o3; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	o3	70.1%	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 MiMo-V2-Omni; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	MiMo-V2-Omni	69.9%	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 12B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Gemma 4 12B Unified	69.7%	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 Gemma 4 26B A4B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Gemma 4 26B A4B	69.2%	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; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Grok 4	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 4.1 Opus Thinking; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude 4.1 Opus Thinking	67.9%	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 2.5 Flash; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Gemini 2.5 Flash	65.5%	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-5.4 nano; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.4 nano	65.4%	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 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	64.9%	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 4.1 Fast (Reasoning); bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Grok 4.1 Fast (Reasoning)	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 Command A+; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Command A+	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 Claude 4 Sonnet; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude 4 Sonnet	62.4%	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 Llama 4 Maverick; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Llama 4 Maverick	62.1%	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 Grok 4 Fast (Reasoning); bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Grok 4 Fast (Reasoning)	61.8%	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-4.1; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-4.1	61.2%	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-4.1 mini; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-4.1 mini	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 Small 4; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Mistral Small 4	56.8%	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 Mistral Small 4 (Reasoning); bulk export does not retain a complete upstream harness configuration. mistral-small-4-reasoning	Mistral Small 4	56.8%	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 Mistral Large 3; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Mistral Large 3	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 Gemini 1.5 Pro; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Gemini 1.5 Pro	55%	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	53.2%	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 Mistral Medium 3; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Mistral Medium 3	53%	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	52.9%	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 E4B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Gemma 4 E4B	51.4%	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	48.4%	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 Gemma 3 27B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Gemma 3 27B	48%	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 E2B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Gemma 4 E2B	44.6%	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 Nova Pro; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Nova Pro	44.3%	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-4o mini; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-4o mini	41.5%	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	40.1%	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 Claude 3 Haiku; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude 3 Haiku	30.8%	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 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	26.5%	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 Opus 5; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude Opus 5	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 Gemini 3.5 Flash; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Gemini 3.5 Flash	84.3%	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-5.6 Sol; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.6 Sol	83.4%	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 Gemini 3.6 Flash; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Gemini 3.6 Flash	83.2%	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-5.6 Terra; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.6 Terra	80.7%	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 K3; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Kimi K3	80.5%	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 Muse Spark; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Muse Spark	80.5%	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 Qwen3.7 Plus; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Qwen3.7-Plus	80.5%	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 4.5; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Grok 4.5	80.4%	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-5.5; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.5	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 Kimi K2.6; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Kimi K2.6	79.4%	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 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	79%	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 (Adaptive); bulk export does not retain a complete upstream harness configuration. claude-opus-4-7-max	Claude Opus 4.7	78.8%	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 Gemini 3 Flash; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Gemini 3 Flash	78.6%	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-5.6 Luna; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.6 Luna	78.6%	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 MiniMax M3; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	MiniMax M3	78.6%	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-5.3 Codex; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.3-Codex	78.5%	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-5.4; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.4	78.4%	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 4.3; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Grok 4.3	78.1%	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 Qwen3.6 Plus; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Qwen3.6 Plus	78%	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 5; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude Sonnet 5	77.3%	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; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Qwen3.5 397B A17B	77.3%	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	77.3%	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 4.7; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude Opus 4.7	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 GPT-5.2-Codex; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.2 Codex	76.3%	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 3.1 Flash-Lite; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Gemini 3.1 Flash-Lite	75.5%	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-5.1; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.1	75.5%	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 Opus 4.6 (Adaptive); bulk export does not retain a complete upstream harness configuration. claude-opus-4-6-max	Claude Opus 4.6	75.4%	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 Kimi K2.5; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Kimi K2.5	75.4%	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 Kimi K2.5 (Reasoning); bulk export does not retain a complete upstream harness configuration. kimi-k2-5-thinking	Kimi K2.5	75.4%	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 Step 3.7 Flash; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Step 3.7 Flash	75.3%	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 Qwen3.5-122B-A10B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Qwen3.5-122B-A10B	75%	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	75%	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	75%	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	74.9%	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 Qwen3.6-27B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Qwen3.6 27B	74.6%	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-5 (medium); bulk export does not retain a complete upstream harness configuration. gpt-5-medium	GPT-5	74.3%	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-5 (high); bulk export does not retain a complete upstream harness configuration. gpt-5-high	GPT-5	74.2%	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 Thinking; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude Opus 4.5 Thinking	74%	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 Inkling; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Inkling	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 Gemma 4 31B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Gemma 4 31B	73.4%	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-5.4 mini; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.4 mini	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 GLM-5V-Turbo; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GLM-5V-Turbo	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 Qwen3.5-35B-A3B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Qwen3.5-35B-A3B	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 Claude Opus 4.6; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude Opus 4.6	72.5%	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-5.1-Codex; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.1-Codex	72.5%	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-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	72.5%	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 Opus 4.5; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude Opus 4.5	71.2%	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 Sonnet 4.6; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude Sonnet 4.6	70.6%	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 o3; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	o3	70.1%	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 MiMo-V2-Omni; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	MiMo-V2-Omni	69.9%	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 12B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Gemma 4 12B Unified	69.7%	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 Gemma 4 26B A4B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Gemma 4 26B A4B	69.2%	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; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Grok 4	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 4.1 Opus Thinking; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude 4.1 Opus Thinking	67.9%	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 2.5 Flash; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Gemini 2.5 Flash	65.5%	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-5.4 nano; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.4 nano	65.4%	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 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	64.9%	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 4.1 Fast (Reasoning); bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Grok 4.1 Fast (Reasoning)	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 Command A+; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Command A+	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 Claude 4 Sonnet; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude 4 Sonnet	62.4%	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 Llama 4 Maverick; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Llama 4 Maverick	62.1%	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 Grok 4 Fast (Reasoning); bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Grok 4 Fast (Reasoning)	61.8%	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-4.1; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-4.1	61.2%	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-4.1 mini; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-4.1 mini	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 Small 4; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Mistral Small 4	56.8%	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 Mistral Small 4 (Reasoning); bulk export does not retain a complete upstream harness configuration. mistral-small-4-reasoning	Mistral Small 4	56.8%	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 Mistral Large 3; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Mistral Large 3	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 Gemini 1.5 Pro; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Gemini 1.5 Pro	55%	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	53.2%	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 Mistral Medium 3; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Mistral Medium 3	53%	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	52.9%	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 E4B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Gemma 4 E4B	51.4%	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	48.4%	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 Gemma 3 27B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Gemma 3 27B	48%	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 E2B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Gemma 4 E2B	44.6%	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 Nova Pro; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Nova Pro	44.3%	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-4o mini; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-4o mini	41.5%	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	40.1%	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 Claude 3 Haiku; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude 3 Haiku	30.8%	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 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	26.5%	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 3.5 Flash; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Gemini 3.5 Flash	84.3%	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-5.6 Sol; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.6 Sol	83.4%	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 Gemini 3.6 Flash; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Gemini 3.6 Flash	83.2%	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
Artificial Analysis MMMU-Pro independent evaluation. source label without registered configuration ID	Gemini 3.6 Flash	83.2%	CurrentArtificial Analysis evaluation harness	Score input source-checked	Gemini 3.6 Flash (Artificial Analysis) ↗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	80.7%	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 K3; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Kimi K3	80.5%	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 Muse Spark; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Muse Spark	80.5%	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 Qwen3.7 Plus; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Qwen3.7-Plus	80.5%	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 4.5; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Grok 4.5	80.4%	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-5.5; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.5	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 Kimi K2.6; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Kimi K2.6	79.4%	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 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	79%	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 MMMU-Pro independent evaluation. source label without registered configuration ID	Gemini 3.5 Flash-Lite	79%	CurrentArtificial Analysis evaluation harness	Score input source-checked	Gemini 3.5 Flash-Lite (Artificial Analysis) ↗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	78.8%	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 Gemini 3 Flash; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Gemini 3 Flash	78.6%	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-5.6 Luna; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.6 Luna	78.6%	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 MiniMax M3; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	MiniMax M3	78.6%	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-5.3 Codex; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.3-Codex	78.5%	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 GPT-5.4; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.4	78.4%	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 4.3; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Grok 4.3	78.1%	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 Qwen3.6 Plus; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Qwen3.6 Plus	78%	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 5; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude Sonnet 5	77.3%	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; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Qwen3.5 397B A17B	77.3%	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	77.3%	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 4.7; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude Opus 4.7	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 GPT-5.2-Codex; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.2 Codex	76.3%	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 3.1 Flash-Lite; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Gemini 3.1 Flash-Lite	75.5%	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 GPT-5.1; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.1	75.5%	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 Opus 4.6 (Adaptive); bulk export does not retain a complete upstream harness configuration. claude-opus-4-6-max	Claude Opus 4.6	75.4%	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 Kimi K2.5; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Kimi K2.5	75.4%	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 Kimi K2.5 (Reasoning); bulk export does not retain a complete upstream harness configuration. kimi-k2-5-thinking	Kimi K2.5	75.4%	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 Step 3.7 Flash; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Step 3.7 Flash	75.3%	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 Qwen3.5-122B-A10B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Qwen3.5-122B-A10B	75%	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	75%	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	Qwen3.6-35B-A3B	75%	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	Gemini 2.5 Pro	74.9%	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 Qwen3.6-27B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Qwen3.6 27B	74.6%	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-5 (medium); bulk export does not retain a complete upstream harness configuration. gpt-5-medium	GPT-5	74.3%	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-5 (high); bulk export does not retain a complete upstream harness configuration. gpt-5-high	GPT-5	74.2%	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 Thinking; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude Opus 4.5 Thinking	74%	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 Inkling; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Inkling	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 Gemma 4 31B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Gemma 4 31B	73.4%	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-5.4 mini; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.4 mini	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 GLM-5V-Turbo; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GLM-5V-Turbo	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 Qwen3.5-35B-A3B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Qwen3.5-35B-A3B	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 Claude Opus 4.6; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude Opus 4.6	72.5%	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 GPT-5.1-Codex; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.1-Codex	72.5%	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 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	72.5%	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 Opus 4.5; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude Opus 4.5	71.2%	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 Sonnet 4.6; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude Sonnet 4.6	70.6%	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 o3; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	o3	70.1%	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 MiMo-V2-Omni; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	MiMo-V2-Omni	69.9%	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 12B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Gemma 4 12B Unified	69.7%	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 Gemma 4 26B A4B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Gemma 4 26B A4B	69.2%	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 Grok 4; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Grok 4	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 4.1 Opus Thinking; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude 4.1 Opus Thinking	67.9%	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 2.5 Flash; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Gemini 2.5 Flash	65.5%	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 GPT-5.4 nano; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.4 nano	65.4%	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 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	64.9%	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 4.1 Fast (Reasoning); bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Grok 4.1 Fast (Reasoning)	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 Command A+; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Command A+	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 Claude 4 Sonnet; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude 4 Sonnet	62.4%	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 Llama 4 Maverick; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Llama 4 Maverick	62.1%	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 Grok 4 Fast (Reasoning); bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Grok 4 Fast (Reasoning)	61.8%	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-4.1; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-4.1	61.2%	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-4.1 mini; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-4.1 mini	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 Small 4; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Mistral Small 4	56.8%	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 Mistral Small 4 (Reasoning); bulk export does not retain a complete upstream harness configuration. mistral-small-4-reasoning	Mistral Small 4	56.8%	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 Mistral Large 3; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Mistral Large 3	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 Gemini 1.5 Pro; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Gemini 1.5 Pro	55%	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	Nemotron 3 Nano Omni 30B A3B	53.2%	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 Mistral Medium 3; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Mistral Medium 3	53%	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	Llama 4 Scout	52.9%	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 E4B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Gemma 4 E4B	51.4%	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	Grok 4.1 Fast	48.4%	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 Gemma 3 27B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Gemma 3 27B	48%	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 E2B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Gemma 4 E2B	44.6%	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 Nova Pro; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Nova Pro	44.3%	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-4o mini; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-4o mini	41.5%	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 GPT-4.1 nano; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-4.1 nano	40.1%	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 Claude 3 Haiku; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude 3 Haiku	30.8%	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 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	26.5%	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
MMMU-Pro official protocol; original input order; images prepended to text; max reasoning; average of three runs. kimi-k3-max	Kimi K3	81.6%	CurrentProvider evaluation harness	Public reference · not scored provider-reported	Kimi K3: Open Frontier Intelligence ↗Observed 2026-07-16 · checked 2026-07-21
Exact model variant as listed on the BenchLM public benchmark leaderboard page. source label without registered configuration ID	Gemini 3.1 Pro Preview	83.9%	CurrentBenchLM aggregated public evaluation	Source carrier · no additional scoring weight source-checked	BenchLM public benchmark leaderboards ↗Observed 2026-07-15 · checked 2026-07-15
Artificial Analysis public model evaluation page; exact provider variant as listed on the page. source label without registered configuration ID	Gemini 3.5 Flash	83.9%	CurrentArtificial Analysis evaluation harness	Score input source-checked	Artificial Analysis evaluations for gemini-3-5-flash-medium ↗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	Gemini 3.5 Flash	83.6%	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	GPT-5.6 Sol	83.4%	CurrentArtificial Analysis evaluation harness	Score input source-checked	Artificial Analysis evaluations for gpt-5-6-sol ↗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	GPT-5.6 Sol	83%	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	Gemini 3.1 Pro Preview	82.4%	CurrentArtificial Analysis evaluation harness	Score input 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	GPT-5.4	81.2%	CurrentBenchLM aggregated public evaluation	Source carrier · no additional scoring weight source-checked	BenchLM public benchmark leaderboards ↗Observed 2026-07-15 · checked 2026-07-15
Exact model variant as listed on the BenchLM public benchmark leaderboard page. source label without registered configuration ID	GPT-5.5	81.2%	CurrentBenchLM aggregated public evaluation	Source carrier · no additional scoring weight source-checked	BenchLM public benchmark leaderboards ↗Observed 2026-07-15 · checked 2026-07-15
Exact model variant as listed on the BenchLM public benchmark leaderboard page. source label without registered configuration ID	GPT-5.6 Terra	80.7%	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	GPT-5.6 Terra	80.7%	CurrentArtificial Analysis evaluation harness	Score input source-checked	Artificial Analysis evaluations for gpt-5-6-terra ↗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	Grok 4.5	80.4%	CurrentArtificial Analysis evaluation harness	Score input source-checked	Artificial Analysis evaluations for grok-4-5 ↗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	Grok 4.5	80.4%	CurrentBenchLM aggregated public evaluation	Source carrier · no additional scoring weight source-checked	BenchLM public benchmark leaderboards ↗Observed 2026-07-15 · checked 2026-07-15
Exact model variant as listed on the BenchLM public benchmark leaderboard page. source label without registered configuration ID	Muse Spark	80.4%	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	GPT-5.5	79.9%	CurrentArtificial Analysis evaluation harness	Score input source-checked	Artificial Analysis evaluations for gpt-5-5 ↗Observed 2026-07-15 · checked 2026-07-15
Kimi K2.6 source label without registered configuration ID	Kimi K2.6	79.4%	CurrentArtificial Analysis independent evaluation	Score input 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	Qwen3.7-Plus	79%	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	Qwen3.6 Plus	78.8%	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	GPT-5.6 Luna	78.5%	CurrentArtificial Analysis evaluation harness	Score input 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	MiniMax M3	78.5%	CurrentArtificial Analysis evaluation harness	Score input source-checked	Artificial Analysis evaluations for minimax-m3 ↗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	GPT-5.3-Codex	78.5%	CurrentArtificial Analysis evaluation harness	Score input 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	Kimi K2.5	78.5%	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	GPT-5.3-Codex	78.5%	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	GPT-5.4	78.4%	CurrentArtificial Analysis evaluation harness	Score input source-checked	Artificial Analysis evaluations for gpt-5-4 ↗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	GPT-5.6 Luna	78.4%	CurrentBenchLM aggregated public evaluation	Source carrier · no additional scoring weight source-checked	BenchLM public benchmark leaderboards ↗Observed 2026-07-15 · checked 2026-07-15
Exact model variant as listed on the BenchLM public benchmark leaderboard page. source label without registered configuration ID	MiniMax M3	78.1%	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	Grok 4.3	78.1%	CurrentBenchLM aggregated public evaluation	Source carrier · no additional scoring weight source-checked	BenchLM public benchmark leaderboards ↗Observed 2026-07-15 · checked 2026-07-15
Artificial Analysis public model evaluation page; exact provider variant as listed on the page. source label without registered configuration ID	Grok 4.3	78.1%	CurrentArtificial Analysis evaluation harness	Score input 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	Claude Sonnet 5	77.3%	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	Claude Opus 4.6	77.3%	CurrentBenchLM aggregated public evaluation	Source carrier · no additional scoring weight source-checked	BenchLM public benchmark leaderboards ↗Observed 2026-07-15 · checked 2026-07-15
Qwen3.5 397B A17B (Reasoning) source label without registered configuration ID	Qwen3.5 397B A17B	77.3%	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	Claude Sonnet 5	77.3%	CurrentArtificial Analysis evaluation harness	Score input 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	Gemma 4 31B	76.9%	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	GPT-5.2 Codex	76.3%	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	Gemini 3.1 Flash-Lite	75.5%	CurrentBenchLM aggregated public evaluation	Source carrier · no additional scoring weight source-checked	BenchLM public benchmark leaderboards ↗Observed 2026-07-15 · checked 2026-07-15
MiMo-V2.5 source label without registered configuration ID	MiMo-V2.5	75.4%	CurrentArtificial Analysis independent evaluation	Score input source-checked	Artificial Analysis Models Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15
Qwen3.5 27B (Reasoning) source label without registered configuration ID	Qwen3.5 27B	75.0%	CurrentArtificial Analysis independent evaluation	Score input 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	75.0%	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	Gemini 2.5 Pro	74.9%	CurrentArtificial Analysis independent evaluation	Score input source-checked	Artificial Analysis Models Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15
Qwen3.6 27B (Reasoning) source label without registered configuration ID	Qwen3.6 27B	74.6%	CurrentArtificial Analysis independent evaluation	Score input 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	Grok 4.20	74.6%	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	GPT-5	74.2%	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	73.8%	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 mini (xhigh) gpt-5-4-mini-xhigh	GPT-5.4 mini	73.3%	CurrentArtificial Analysis independent evaluation	Public reference · not scored 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	Gemini 2.5 Flash	73.1%	CurrentArtificial Analysis independent evaluation	Score input 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	72.7%	CurrentArtificial Analysis independent evaluation	Score input 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	Claude Opus 4.5	70.6%	CurrentBenchLM aggregated public evaluation	Source carrier · no additional scoring weight source-checked	BenchLM public benchmark leaderboards ↗Observed 2026-07-15 · checked 2026-07-15
Qwen3.5 Omni Plus source label without registered configuration ID	Qwen3.5 Omni Plus	70.5%	CurrentArtificial Analysis independent evaluation	Score input source-checked	Artificial Analysis Models Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15
o3 source label without registered configuration ID	o3	70.1%	CurrentArtificial Analysis independent evaluation	Score input source-checked	Artificial Analysis Models Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15
Gemma 4 12B (Reasoning) source label without registered configuration ID	Gemma 4 12B Unified	69.7%	CurrentArtificial Analysis independent evaluation	Score input 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	69.2%	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	o4-mini	69.2%	CurrentArtificial Analysis independent evaluation	Public reference · not scored source-checked	Artificial Analysis Models Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15
GPT-5 mini (medium) gpt-5-mini-medium	GPT-5 mini	68.8%	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	Grok 4	68.8%	CurrentArtificial Analysis independent evaluation	Score input 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	Claude Sonnet 4.5	68.7%	CurrentArtificial Analysis independent evaluation	Score input 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	Claude Opus 4.1	67.9%	CurrentArtificial Analysis independent evaluation	Score input 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	GPT-5.4 nano	66.1%	CurrentBenchLM aggregated public evaluation	Source carrier · no additional scoring weight source-checked	BenchLM public benchmark leaderboards ↗Observed 2026-07-15 · checked 2026-07-15
Artificial Analysis public model evaluation page; exact provider variant as listed on the page. source label without registered configuration ID	Mistral Medium 3.5	64.9%	CurrentArtificial Analysis evaluation harness	Score input source-checked	Artificial Analysis evaluations for mistral-medium-3-5 ↗Observed 2026-07-15 · checked 2026-07-15
Nova 2.0 Pro Preview (medium) source label without registered configuration ID	Nova 2 Pro	64.5%	CurrentArtificial Analysis independent evaluation	Score input source-checked	Artificial Analysis Models Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15
Command A+ source label without registered configuration ID	Command A+	63.2%	CurrentArtificial Analysis independent evaluation	Score input source-checked	Artificial Analysis Models Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15
Claude 4 Sonnet (Reasoning) source label without registered configuration ID	Claude Sonnet 4	61.8%	CurrentArtificial Analysis independent evaluation	Score input source-checked	Artificial Analysis Models Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15
Grok 4 Fast (Reasoning) source label without registered configuration ID	Grok 4 Fast	61.8%	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	Mistral Small 4	56.8%	CurrentArtificial Analysis evaluation harness	Score input source-checked	Artificial Analysis evaluations for mistral-small-4 ↗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	Mistral Small 4	56.8%	CurrentBenchLM aggregated public evaluation	Source carrier · no additional scoring weight source-checked	BenchLM public benchmark leaderboards ↗Observed 2026-07-15 · checked 2026-07-15
Exact model variant as listed on the BenchLM public benchmark leaderboard page. source label without registered configuration ID	Mistral Large 3	55.7%	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	Mistral Large 3	55.7%	CurrentArtificial Analysis evaluation harness	Score input 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	Claude Haiku 4.5	55.1%	CurrentArtificial Analysis evaluation harness	Score input source-checked	Artificial Analysis evaluations for claude-4-5-haiku ↗Observed 2026-07-15 · checked 2026-07-15
Exact model variant and provider evaluation configuration stated in the Gemini 3.5 Flash model card; single-attempt where specified. source label without registered configuration ID	Gemini 3.5 Flash	83.6%	CurrentProvider evaluation harness	Score input provider-reported	Gemini 3.5 Flash model card ↗Observed 2026-05-19 · checked 2026-07-15
Exact model variant and provider evaluation configuration stated in the Gemini 3.5 Flash model card; single-attempt where specified. source label without registered configuration ID	GPT-5.5	81.2%	CurrentProvider evaluation harness	Public reference · not scored provider-reported	Gemini 3.5 Flash model card ↗Observed 2026-05-19 · checked 2026-07-15
Exact model variant and provider evaluation configuration stated in the Gemini 3.5 Flash model card; single-attempt where specified. source label without registered configuration ID	Gemini 3.1 Pro Preview	80.5%	CurrentProvider evaluation harness	Public reference · not scored provider-reported	Gemini 3.5 Flash model card ↗Observed 2026-05-19 · checked 2026-07-15
GPT-4o mini source label without registered configuration ID	GPT-4o mini	37.6%	2024MMMU-Pro official owner evaluation (paper / leaderboard_data.json unlabeled row)	Score input official-leaderboard	MMMU permanent refresh source ↗Observed 2024-07-18 · checked 2026-08-18
Gemini 1.5 Pro (0523) source label without registered configuration ID	Gemini 1.5 Pro (May '24)	43.5%	2024MMMU-Pro official owner evaluation (paper / leaderboard_data.json unlabeled row)	Score input official-leaderboard	MMMU permanent refresh source ↗Observed 2024-05-23 · checked 2026-08-18
GPT-4o (0513) source label without registered configuration ID	GPT-4o (May '24)	51.9%	2024MMMU-Pro official owner evaluation (paper / leaderboard_data.json unlabeled row)	Score input official-leaderboard	MMMU permanent refresh source ↗Observed 2024-05-13 · checked 2026-08-18

Exact published model / configuration

Canonical product

Score

Version / system

Evidence

Source / observed

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

muse-glimmer-30b-high

Muse Glimmer 30B

74%

1,730-question setArtificial Analysis MMMU-Pro documented evaluation

Public reference · not scored

provider-reported

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

Qwen3.8 Max (xhigh, default reasoning effort)

qwen-3-8-max-xhigh

Qwen3.8 Max

82.3%

CurrentQwen provider evaluation

Public reference · not scored

provider-reported

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

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

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 Gemini 3.5 Flash; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

84.3%

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-5.6 Sol; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

83.4%

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 Gemini 3.6 Flash; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

83.2%

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-5.6 Terra; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

80.7%

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 K3; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

80.5%

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 Muse Spark; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

80.5%

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 Qwen3.7 Plus; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

80.5%

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 4.5; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

80.4%

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-5.5; 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 Kimi K2.6; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

79.4%

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

source label without registered configuration ID

79%

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 (Adaptive); bulk export does not retain a complete upstream harness configuration.

claude-opus-4-7-max

78.8%

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 Gemini 3 Flash; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

78.6%

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-5.6 Luna; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

78.6%

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 MiniMax M3; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

78.6%

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-5.3 Codex; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

78.5%

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-5.4; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

78.4%

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 4.3; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

78.1%

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 Qwen3.6 Plus; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

78%

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 5; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

77.3%

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; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

77.3%

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

77.3%

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 4.7; 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 GPT-5.2-Codex; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

76.3%

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 3.1 Flash-Lite; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

75.5%

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-5.1; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

75.5%

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 Opus 4.6 (Adaptive); bulk export does not retain a complete upstream harness configuration.

claude-opus-4-6-max

75.4%

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 Kimi K2.5; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

75.4%

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

kimi-k2-5-thinking

75.4%

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 Step 3.7 Flash; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

75.3%

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

source label without registered configuration ID

75%

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

75%

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

75%

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

74.9%

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

source label without registered configuration ID

74.6%

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-5 (medium); bulk export does not retain a complete upstream harness configuration.

gpt-5-medium

74.3%

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-5 (high); bulk export does not retain a complete upstream harness configuration.

gpt-5-high

74.2%

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 Thinking; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

74%

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

source label without registered configuration ID

73.4%

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-5.4 mini; 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 GLM-5V-Turbo; 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 Qwen3.5-35B-A3B; 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 Claude Opus 4.6; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

72.5%

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-5.1-Codex; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

72.5%

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-5.1-Codex-Max; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

72.5%

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 Opus 4.5; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

71.2%

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 Sonnet 4.6; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

70.6%

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 o3; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

70.1%

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

source label without registered configuration ID

69.9%

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 12B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

69.7%

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

source label without registered configuration ID

69.2%

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; 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 4.1 Opus Thinking; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

67.9%

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 2.5 Flash; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

65.5%

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-5.4 nano; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

65.4%

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

source label without registered configuration ID

64.9%

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 4.1 Fast (Reasoning); 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 Command A+; 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 Claude 4 Sonnet; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

62.4%

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 Llama 4 Maverick; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

62.1%

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

source label without registered configuration ID

61.8%

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-4.1; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

61.2%

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-4.1 mini; 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 Small 4; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

56.8%

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

mistral-small-4-reasoning

56.8%

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 Mistral Large 3; 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 Gemini 1.5 Pro; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

55%

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

53.2%

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 Mistral Medium 3; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

53%

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

52.9%

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 E4B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

51.4%

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

48.4%

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

source label without registered configuration ID

48%

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 E2B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

44.6%

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 Nova Pro; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

44.3%

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-4o mini; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

41.5%

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

40.1%

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 Claude 3 Haiku; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

30.8%

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

source label without registered configuration ID

26.5%

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 Opus 5; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

Claude Opus 5

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 Gemini 3.5 Flash; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

84.3%

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-5.6 Sol; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

83.4%

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 Gemini 3.6 Flash; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

83.2%

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-5.6 Terra; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

80.7%

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 K3; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

80.5%

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 Muse Spark; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

80.5%

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 Qwen3.7 Plus; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

80.5%

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 4.5; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

80.4%

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-5.5; 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 Kimi K2.6; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

79.4%

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

source label without registered configuration ID

79%

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 (Adaptive); bulk export does not retain a complete upstream harness configuration.

claude-opus-4-7-max

78.8%

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 Gemini 3 Flash; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

78.6%

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-5.6 Luna; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

78.6%

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 MiniMax M3; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

78.6%

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-5.3 Codex; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

78.5%

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-5.4; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

78.4%

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 4.3; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

78.1%

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 Qwen3.6 Plus; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

78%

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 5; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

77.3%

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; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

77.3%

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

77.3%

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 4.7; 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 GPT-5.2-Codex; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

76.3%

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 3.1 Flash-Lite; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

75.5%

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-5.1; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

75.5%

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 Opus 4.6 (Adaptive); bulk export does not retain a complete upstream harness configuration.

claude-opus-4-6-max

75.4%

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 Kimi K2.5; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

75.4%

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

kimi-k2-5-thinking

75.4%

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 Step 3.7 Flash; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

75.3%

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

source label without registered configuration ID

75%

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

75%

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

75%

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

74.9%

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

source label without registered configuration ID

74.6%

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-5 (medium); bulk export does not retain a complete upstream harness configuration.

gpt-5-medium

74.3%

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-5 (high); bulk export does not retain a complete upstream harness configuration.

gpt-5-high

74.2%

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 Thinking; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

74%

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 Inkling; 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 Gemma 4 31B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

73.4%

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-5.4 mini; 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 GLM-5V-Turbo; 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 Qwen3.5-35B-A3B; 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 Claude Opus 4.6; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

72.5%

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-5.1-Codex; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

72.5%

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-5.1-Codex-Max; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

72.5%

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 Opus 4.5; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

71.2%

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 Sonnet 4.6; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

70.6%

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 o3; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

70.1%

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

source label without registered configuration ID

69.9%

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 12B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

69.7%

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

source label without registered configuration ID

69.2%

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; 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 4.1 Opus Thinking; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

67.9%

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 2.5 Flash; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

65.5%

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-5.4 nano; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

65.4%

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

source label without registered configuration ID

64.9%

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 4.1 Fast (Reasoning); 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 Command A+; 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 Claude 4 Sonnet; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

62.4%

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 Llama 4 Maverick; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

62.1%

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

source label without registered configuration ID

61.8%

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-4.1; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

61.2%

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-4.1 mini; 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 Small 4; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

56.8%

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

mistral-small-4-reasoning

56.8%

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 Mistral Large 3; 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 Gemini 1.5 Pro; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

55%

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

53.2%

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 Mistral Medium 3; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

53%

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

52.9%

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 E4B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

51.4%

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

48.4%

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

source label without registered configuration ID

48%

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 E2B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

44.6%

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 Nova Pro; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

44.3%

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-4o mini; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

41.5%

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

40.1%

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 Claude 3 Haiku; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

30.8%

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

source label without registered configuration ID

26.5%

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 3.5 Flash; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

84.3%

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-5.6 Sol; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

83.4%

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 Gemini 3.6 Flash; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

83.2%

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

Artificial Analysis MMMU-Pro independent evaluation.

source label without registered configuration ID

83.2%

CurrentArtificial Analysis evaluation harness

Score input

source-checked

Gemini 3.6 Flash (Artificial Analysis) ↗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

80.7%

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 K3; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

80.5%

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 Muse Spark; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

80.5%

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 Qwen3.7 Plus; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

80.5%

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 4.5; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

80.4%

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-5.5; 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 Kimi K2.6; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

79.4%

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

source label without registered configuration ID

79%

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 MMMU-Pro independent evaluation.

source label without registered configuration ID

79%

CurrentArtificial Analysis evaluation harness

Score input

source-checked

Gemini 3.5 Flash-Lite (Artificial Analysis) ↗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

78.8%

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 Gemini 3 Flash; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

78.6%

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-5.6 Luna; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

78.6%

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 MiniMax M3; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

78.6%

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-5.3 Codex; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

78.5%

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 GPT-5.4; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

78.4%

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 4.3; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

78.1%

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 Qwen3.6 Plus; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

78%

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 5; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

77.3%

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; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

77.3%

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

77.3%

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 4.7; 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 GPT-5.2-Codex; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

76.3%

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 3.1 Flash-Lite; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

75.5%

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 GPT-5.1; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

75.5%

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 Opus 4.6 (Adaptive); bulk export does not retain a complete upstream harness configuration.

claude-opus-4-6-max

75.4%

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 Kimi K2.5; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

75.4%

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

kimi-k2-5-thinking

75.4%

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 Step 3.7 Flash; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

75.3%

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

source label without registered configuration ID

75%

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

75%

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

75%

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

74.9%

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

source label without registered configuration ID

74.6%

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-5 (medium); bulk export does not retain a complete upstream harness configuration.

gpt-5-medium

74.3%

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-5 (high); bulk export does not retain a complete upstream harness configuration.

gpt-5-high

74.2%

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 Thinking; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

74%

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 Inkling; 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 Gemma 4 31B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

73.4%

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-5.4 mini; 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 GLM-5V-Turbo; 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 Qwen3.5-35B-A3B; 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 Claude Opus 4.6; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

72.5%

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

source label without registered configuration ID

72.5%

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

source label without registered configuration ID

72.5%

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 Opus 4.5; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

71.2%

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 Sonnet 4.6; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

70.6%

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 o3; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

70.1%

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

source label without registered configuration ID

69.9%

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 12B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

69.7%

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

source label without registered configuration ID

69.2%

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 Grok 4; 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 4.1 Opus Thinking; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

67.9%

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 2.5 Flash; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

65.5%

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

source label without registered configuration ID

65.4%

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

source label without registered configuration ID

64.9%

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 4.1 Fast (Reasoning); 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 Command A+; 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 Claude 4 Sonnet; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

62.4%

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 Llama 4 Maverick; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

62.1%

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

source label without registered configuration ID

61.8%

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-4.1; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

61.2%

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-4.1 mini; 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 Small 4; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

56.8%

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

mistral-small-4-reasoning

56.8%

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 Mistral Large 3; 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 Gemini 1.5 Pro; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

55%

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

53.2%

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 Mistral Medium 3; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

53%

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

52.9%

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 E4B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

51.4%

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

48.4%

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

source label without registered configuration ID

48%

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 E2B; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

44.6%

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 Nova Pro; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

44.3%

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-4o mini; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

41.5%

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

source label without registered configuration ID

40.1%

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 Claude 3 Haiku; bulk export does not retain a complete upstream harness configuration.

source label without registered configuration ID

30.8%

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

source label without registered configuration ID

26.5%

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

MMMU-Pro official protocol; original input order; images prepended to text; max reasoning; average of three runs.

kimi-k3-max

81.6%

CurrentProvider evaluation harness

Public reference · not scored

provider-reported

Kimi K3: Open Frontier Intelligence ↗Observed 2026-07-16 · checked 2026-07-21

Exact model variant as listed on the BenchLM public benchmark leaderboard page.

source label without registered configuration ID

83.9%

CurrentBenchLM aggregated public evaluation

Source carrier · no additional scoring weight

source-checked

BenchLM public benchmark leaderboards ↗Observed 2026-07-15 · checked 2026-07-15

Artificial Analysis public model evaluation page; exact provider variant as listed on the page.

source label without registered configuration ID

83.9%

CurrentArtificial Analysis evaluation harness

Score input

source-checked

Artificial Analysis evaluations for gemini-3-5-flash-medium ↗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

83.6%

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

83.4%

CurrentArtificial Analysis evaluation harness

Score input

source-checked

Artificial Analysis evaluations for gpt-5-6-sol ↗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

83%

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

82.4%

CurrentArtificial Analysis evaluation harness

Score input

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

81.2%

CurrentBenchLM aggregated public evaluation

Source carrier · no additional scoring weight

source-checked

BenchLM public benchmark leaderboards ↗Observed 2026-07-15 · checked 2026-07-15

Exact model variant as listed on the BenchLM public benchmark leaderboard page.

source label without registered configuration ID

81.2%

CurrentBenchLM aggregated public evaluation

Source carrier · no additional scoring weight

source-checked

BenchLM public benchmark leaderboards ↗Observed 2026-07-15 · checked 2026-07-15

Exact model variant as listed on the BenchLM public benchmark leaderboard page.

source label without registered configuration ID

80.7%

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

80.7%

CurrentArtificial Analysis evaluation harness

Score input

source-checked

Artificial Analysis evaluations for gpt-5-6-terra ↗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

80.4%

CurrentArtificial Analysis evaluation harness

Score input

source-checked

Artificial Analysis evaluations for grok-4-5 ↗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

80.4%

CurrentBenchLM aggregated public evaluation

Source carrier · no additional scoring weight

source-checked

BenchLM public benchmark leaderboards ↗Observed 2026-07-15 · checked 2026-07-15

Exact model variant as listed on the BenchLM public benchmark leaderboard page.

source label without registered configuration ID

80.4%

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

79.9%

CurrentArtificial Analysis evaluation harness

Score input

source-checked

Artificial Analysis evaluations for gpt-5-5 ↗Observed 2026-07-15 · checked 2026-07-15

Kimi K2.6

source label without registered configuration ID

79.4%

CurrentArtificial Analysis independent evaluation

Score input

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

79%

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

78.8%

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

78.5%

CurrentArtificial Analysis evaluation harness

Score input

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

78.5%

CurrentArtificial Analysis evaluation harness

Score input

source-checked

Artificial Analysis evaluations for minimax-m3 ↗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

78.5%

CurrentArtificial Analysis evaluation harness

Score input

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

78.5%

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

78.5%

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

78.4%

CurrentArtificial Analysis evaluation harness

Score input

source-checked

Artificial Analysis evaluations for gpt-5-4 ↗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

78.4%

CurrentBenchLM aggregated public evaluation

Source carrier · no additional scoring weight

source-checked

BenchLM public benchmark leaderboards ↗Observed 2026-07-15 · checked 2026-07-15

Exact model variant as listed on the BenchLM public benchmark leaderboard page.

source label without registered configuration ID

78.1%

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

78.1%

CurrentBenchLM aggregated public evaluation

Source carrier · no additional scoring weight

source-checked

BenchLM public benchmark leaderboards ↗Observed 2026-07-15 · checked 2026-07-15

Artificial Analysis public model evaluation page; exact provider variant as listed on the page.

source label without registered configuration ID

78.1%

CurrentArtificial Analysis evaluation harness

Score input

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

77.3%

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

77.3%

CurrentBenchLM aggregated public evaluation

Source carrier · no additional scoring weight

source-checked

BenchLM public benchmark leaderboards ↗Observed 2026-07-15 · checked 2026-07-15

Qwen3.5 397B A17B (Reasoning)

source label without registered configuration ID

77.3%

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

77.3%

CurrentArtificial Analysis evaluation harness

Score input

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

76.9%

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

76.3%

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

75.5%

CurrentBenchLM aggregated public evaluation

Source carrier · no additional scoring weight

source-checked

BenchLM public benchmark leaderboards ↗Observed 2026-07-15 · checked 2026-07-15

MiMo-V2.5

source label without registered configuration ID

MiMo-V2.5

75.4%

CurrentArtificial Analysis independent evaluation

Score input

source-checked

Artificial Analysis Models Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15

Qwen3.5 27B (Reasoning)

source label without registered configuration ID

75.0%

CurrentArtificial Analysis independent evaluation

Score input

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

75.0%

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

74.9%

CurrentArtificial Analysis independent evaluation

Score input

source-checked

Artificial Analysis Models Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15

Qwen3.6 27B (Reasoning)

source label without registered configuration ID

74.6%

CurrentArtificial Analysis independent evaluation

Score input

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

Grok 4.20

74.6%

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

74.2%

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

73.8%

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 mini (xhigh)

gpt-5-4-mini-xhigh

73.3%

CurrentArtificial Analysis independent evaluation

Public reference · not scored

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

73.1%

CurrentArtificial Analysis independent evaluation

Score input

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

72.7%

CurrentArtificial Analysis independent evaluation

Score input

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

70.6%

CurrentBenchLM aggregated public evaluation

Source carrier · no additional scoring weight

source-checked

BenchLM public benchmark leaderboards ↗Observed 2026-07-15 · checked 2026-07-15

Qwen3.5 Omni Plus

source label without registered configuration ID

Qwen3.5 Omni Plus

70.5%

CurrentArtificial Analysis independent evaluation

Score input

source-checked

Artificial Analysis Models Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15

source label without registered configuration ID

70.1%

CurrentArtificial Analysis independent evaluation

Score input

source-checked

Artificial Analysis Models Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15

Gemma 4 12B (Reasoning)

source label without registered configuration ID

69.7%

CurrentArtificial Analysis independent evaluation

Score input

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

69.2%

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

o4-mini

69.2%

CurrentArtificial Analysis independent evaluation

Public reference · not scored

source-checked

Artificial Analysis Models Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15

GPT-5 mini (medium)

gpt-5-mini-medium

GPT-5 mini

68.8%

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

68.8%

CurrentArtificial Analysis independent evaluation

Score input

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

Claude Sonnet 4.5

68.7%

CurrentArtificial Analysis independent evaluation

Score input

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

Claude Opus 4.1

67.9%

CurrentArtificial Analysis independent evaluation

Score input

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

66.1%

CurrentBenchLM aggregated public evaluation

Source carrier · no additional scoring weight

source-checked

BenchLM public benchmark leaderboards ↗Observed 2026-07-15 · checked 2026-07-15

Artificial Analysis public model evaluation page; exact provider variant as listed on the page.

source label without registered configuration ID

Mistral Medium 3.5

64.9%

CurrentArtificial Analysis evaluation harness

Score input

source-checked

Artificial Analysis evaluations for mistral-medium-3-5 ↗Observed 2026-07-15 · checked 2026-07-15

Nova 2.0 Pro Preview (medium)

source label without registered configuration ID

Nova 2 Pro

64.5%

CurrentArtificial Analysis independent evaluation

Score input

source-checked

Artificial Analysis Models Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15

Command A+

source label without registered configuration ID

63.2%

CurrentArtificial Analysis independent evaluation

Score input

source-checked

Artificial Analysis Models Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15

Claude 4 Sonnet (Reasoning)

source label without registered configuration ID

Claude Sonnet 4

61.8%

CurrentArtificial Analysis independent evaluation

Score input

source-checked

Artificial Analysis Models Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15

Grok 4 Fast (Reasoning)

source label without registered configuration ID

Grok 4 Fast

61.8%

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

56.8%

CurrentArtificial Analysis evaluation harness

Score input

source-checked

Artificial Analysis evaluations for mistral-small-4 ↗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

56.8%

CurrentBenchLM aggregated public evaluation

Source carrier · no additional scoring weight

source-checked

BenchLM public benchmark leaderboards ↗Observed 2026-07-15 · checked 2026-07-15

Exact model variant as listed on the BenchLM public benchmark leaderboard page.

source label without registered configuration ID

55.7%

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

55.7%

CurrentArtificial Analysis evaluation harness

Score input

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

Claude Haiku 4.5

55.1%

CurrentArtificial Analysis evaluation harness

Score input

source-checked

Artificial Analysis evaluations for claude-4-5-haiku ↗Observed 2026-07-15 · checked 2026-07-15

Exact model variant and provider evaluation configuration stated in the Gemini 3.5 Flash model card; single-attempt where specified.

source label without registered configuration ID

83.6%

CurrentProvider evaluation harness

Score input

provider-reported

Gemini 3.5 Flash model card ↗Observed 2026-05-19 · checked 2026-07-15

Exact model variant and provider evaluation configuration stated in the Gemini 3.5 Flash model card; single-attempt where specified.

source label without registered configuration ID

81.2%

CurrentProvider evaluation harness

Public reference · not scored

provider-reported

Gemini 3.5 Flash model card ↗Observed 2026-05-19 · checked 2026-07-15

Exact model variant and provider evaluation configuration stated in the Gemini 3.5 Flash model card; single-attempt where specified.

source label without registered configuration ID

80.5%

CurrentProvider evaluation harness

Public reference · not scored

provider-reported

Gemini 3.5 Flash model card ↗Observed 2026-05-19 · checked 2026-07-15

GPT-4o mini

source label without registered configuration ID