Knowledge · Benchmark profile
A display-only Artificial Analysis factuality metric for the rate of incorrect answers among non-correct responses.
Data verified 27 Jul 2026 · Methodology 1.8.0
Benchmark score on Artificial Analysis Omniscience Hallucination Rate
Command A+ leads at 14.1%, followed by MiniMax M3 (16.1%) and Qwen3.7-Max (22.9%).
10 ranked models · lower is better · labels show rank and score
| Model | Rank | Provider | Score |
|---|---|---|---|
| Command A+ | #1 | Cohere | 14.1% |
| MiniMax M3 | #2 | MiniMax | 16.1% |
| Qwen3.7-Max | #3 | Alibaba Cloud | 22.9% |
| MiMo-V2.5-Pro | #4 | Xiaomi | 24.5% |
| Grok 4.3 | #5 | xAI | 25% |
| Qwen3.7-Plus | #6 | Alibaba Cloud | 25.5% |
| GLM-5.2 | #7 | Z.AI | 28.1% |
| Nemotron 3 Ultra | #8 | NVIDIA | 28.5% |
| GLM-5.1 | #9 | Z.AI | 29.4% |
| MiMo-V2-Pro | #10 | Xiaomi | 29.9% |
One best score per model · lower is better
| Rank | Model | Provider | License | Evidence use | Score |
|---|---|---|---|---|---|
| #1 | Command A+ command-a-plus | Cohere | open | Estimated reference | 14.1% |
| #2 | MiniMax M3 MiniMax-M3 | MiniMax | open | Reference only | 16.1% |
| #3 | Qwen3.7-Max qwen3.7-max | Alibaba Cloud | closed | Reference only | 22.9% |
| #4 | MiMo-V2.5-Pro mimo-v2-5-pro | Xiaomi | closed | Reference only | 24.5% |
| #5 | Grok 4.3 grok-4.3 | xAI | closed | Reference only | 25% |
| #6 | Qwen3.7-Plus qwen3.7-plus | Alibaba Cloud | closed | Reference only | 25.5% |
| #7 | GLM-5.2 glm-5-2 | Z.AI | open | Estimated reference | 28.1% |
| #8 | Nemotron 3 Ultra nemotron-3-ultra-550b | NVIDIA | open | Estimated reference | 28.5% |
| #9 | GLM-5.1 glm-5-1 | Z.AI | open | Reference only | 29.4% |
| #10 | MiMo-V2-Pro mimo-v2-pro | Xiaomi | closed | Reference only | 29.9% |
| #11 | Gemma 4 E4B gemma-4-e4b | open | Estimated reference | 31.3% | |
| #12 | Qwen3.6 Plus qwen3-6-plus | Alibaba Cloud | closed | Reference only | 32% |
| #13 | Gemma 4 E2B gemma-4-e2b | open | Estimated reference | 32.9% | |
| #14 | Gemini 3.5 Flash-Lite gemini-3.5-flash-lite | closed | Reference only | 33.5% | |
| #15 | GLM-5 glm-5 | Z.AI | open | Reference only | 34% |
| #16 | MiniMax M2.7 minimax-m2-7 | MiniMax | open | Reference only | 34.4% |
| #17 | Claude Opus 4.8 claude-opus-4-8 | Anthropic | closed | Reference only | 35.9% |
| #18 | Claude Opus 4.7 (Adaptive) claude-opus-4-7-max | Anthropic | closed | Estimated reference | 36.2% |
| #19 | Claude Sonnet 5 claude-sonnet-5 | Anthropic | closed | Estimated reference | 37.3% |
| #20 | GPT-4o gpt-4o | OpenAI | closed | Reference only | 37.9% |
| #21 | Muse Spark 1.1 muse-spark-1-1 | Meta | closed | Reference only | 38.1% |
| #22 | Kimi K2.6 kimi-k2.6 | Moonshot AI | open | Estimated reference | 39.3% |
| #23 | Claude 4 Sonnet claude-4-sonnet | Anthropic | closed | Reference only | 40.8% |
| #24 | Qwen 3.6 Max (preview) qwen3-6-max-preview | Alibaba Cloud | closed | Reference only | 44.2% |
| #25 | MiMo-V2-Omni mimo-v2-omni | Xiaomi | closed | Reference only | 44.4% |
| #26 | LFM2.5-8B-A1B lfm2-5-8b-a1b | LiquidAI | open | Estimated reference | 47% |
| #27 | Qwen3.6 27B qwen3.6-27b | Alibaba Cloud | open | Estimated reference | 48.3% |
| #28 | Qwen3.6-35B-A3B qwen3-6-35b-a3b | Alibaba Cloud | open | Estimated reference | 49.7% |
| #29 | Gemini 3.1 Pro gemini-3-1-pro | closed | Estimated reference | 49.9% | |
| #30 | Claude Opus 5 claude-opus-5 | Anthropic | closed | Estimated reference | 50.1% |
| #31 | Kimi K3 kimi-k3 | Moonshot AI | closed | Reference only | 50.9% |
| #32 | Llama 3.1 405B llama-3-1-405b | Meta | open | Estimated reference | 51% |
| #33 | GPT-5.1 gpt.5.1 | OpenAI | closed | Estimated reference | 51.3% |
| #34 | Claude Opus 4.7 claude-opus-4-7 | Anthropic | closed | Reference only | 51.9% |
| #35 | Gemini 3.6 Flash gemini-3.6-flash | closed | Reference only | 53.5% | |
| #36 | Grok 4.5 grok-4.5 | xAI | closed | Reference only | 53.5% |
| #37 | Claude Fable 5 claude-fable-5 | Anthropic | closed | Reference only | 54.9% |
| #38 | Claude Opus 4.5 Thinking claude-opus-4-5-thinking | Anthropic | closed | Estimated reference | 59.8% |
| #39 | Gemini 3.5 Flash gemini-3.5-flash | closed | Estimated reference | 60.7% | |
| #40 | Mistral Medium 3 mistral-medium-3 | Mistral AI | closed | Estimated reference | 60.9% |
| #41 | Claude Opus 4.6 (Adaptive) claude-opus-4-6-thinking | Anthropic | closed | Estimated reference | 61.3% |
| #42 | GLM-5-Turbo glm-5-turbo | Z.AI | closed | Reference only | 62.2% |
| #43 | Inkling inkling | Thinking Machines Lab | open | Reference only | 63.1% |
| #44 | Grok 4 grok-4 | xAI | closed | Reference only | 64.2% |
| #45 | Kimi K2.5 kimi-k2.5 | Moonshot AI | open | Reference only | 64.6% |
| #46 | Kimi K2.5 (Reasoning) kimi-k2-5-reasoning | Moonshot AI | closed | Estimated reference | 64.6% |
| #47 | Claude Sonnet 4.6 claude-sonnet-4-6 | Anthropic | closed | Reference only | 65.9% |
| #48 | Grok 4 Fast (Reasoning) grok-4-fast-reasoning | xAI | closed | Reference only | 66% |
| #49 | GLM-4.6 glm-4.6 | Z.AI | open | Reference only | 66.1% |
| #50 | Mistral Small 4 mistral-small-2603 | Mistral AI | open | Estimated reference | 66.8% |
| #51 | Mistral Small 4 (Reasoning) mistral-small-4-reasoning | Mistral AI | open | Estimated reference | 66.8% |
| #52 | Mistral Large 2 mistral-large-2 | Mistral AI | closed | Estimated reference | 67.8% |
| #53 | GLM-5V-Turbo glm-5v-turbo | Z.AI | closed | Reference only | 67.9% |
| #54 | o1 o1 | OpenAI | closed | Estimated reference | 69.3% |
| #55 | Grok 4.1 Fast (Reasoning) grok-4-1-fast-reasoning | xAI | closed | Reference only | 72.4% |
| #56 | GPT-5.2 Codex gpt-5.2-codex | OpenAI | closed | Reference only | 72.8% |
| #57 | Hy3 hy3 | Tencent | open | Estimated reference | 73% |
| #58 | Hy3 Preview hy3-preview | Tencent | open | Estimated reference | 73% |
| #59 | Muse Spark muse-spark | Meta | closed | Reference only | 73.2% |
| #60 | GPT-5.4 nano gpt.5.4.nano | OpenAI | closed | Reference only | 73.6% |
| #61 | Kimi K2 kimi-k2 | Moonshot AI | closed | Reference only | 74.2% |
| #62 | GPT-5.1-Codex gpt-5-1-codex | OpenAI | closed | Estimated reference | 74.4% |
| #63 | GPT-5.1-Codex-Max gpt-5-1-codex-max | OpenAI | closed | Estimated reference | 74.4% |
| #64 | MiMo-V2-Flash mimo-v2-flash | Xiaomi | open | Reference only | 75.1% |
| #65 | Claude Opus 4.5 claude-opus-4-5 | Anthropic | closed | Reference only | 75.4% |
| #66 | Claude Opus 4.6 claude-opus-4-6 | Anthropic | closed | Reference only | 76% |
| #67 | Granite-4.0-350M granite-4-0-350m | IBM | open | Estimated reference | 77.8% |
| #68 | Nova Pro nova-pro | Amazon | closed | Reference only | 77.9% |
| #69 | Claude 3 Haiku claude-3-haiku | Anthropic | closed | Reference only | 78.2% |
| #70 | Llama 4 Scout llama-4-scout | Meta | open | Reference only | 78.3% |
| #71 | Grok Code Fast 1 grok-code-fast-1 | xAI | closed | Reference only | 78.5% |
| #72 | GPT-4.1 gpt-4-1 | OpenAI | closed | Reference only | 79.6% |
| #73 | GPT-5.2 gpt-5.2 | OpenAI | closed | Estimated reference | 79.7% |
| #74 | Qwen3.5 27B qwen3.5-27b | Alibaba Cloud | open | Reference only | 79.7% |
| #75 | GPT-5 (medium) gpt-5-medium | OpenAI | closed | Reference only | 80.1% |
| #76 | DeepSeek V3.1 (Reasoning) deepseek-v3-1-reasoning | DeepSeek | open | Reference only | 80.3% |
| #77 | Kimi K2.7 Code kimi-k2.7-code | Moonshot AI | open | Estimated reference | 80.3% |
| #78 | GPT-4.1 nano gpt-4-1-nano | OpenAI | closed | Estimated reference | 80.4% |
| #79 | Phi-4 phi-4 | Microsoft | open | Reference only | 80.5% |
| #80 | Gemma 4 12B gemma-4-12b | open | Estimated reference | 80.8% | |
| #81 | Gemma 4 26B A4B gemma-4-26b-a4b | open | Reference only | 80.9% | |
| #82 | Exaone 4.0 32B exaone-4-0-32b | LG AI Research | open | Estimated reference | 81% |
| #83 | Gemini 3.1 Flash-Lite gemini-3.1-flash-lite | closed | Reference only | 81.6% | |
| #84 | Gemma 4 31B gemma-4-31b | open | Reference only | 81.6% | |
| #85 | Nemotron Ultra 253B nemotron-ultra-253b | NVIDIA | open | Estimated reference | 81.7% |
| #86 | Grok 4.1 Fast grok-4-1-fast | xAI | closed | Reference only | 81.8% |
| #87 | GPT-4.1 mini gpt-4-1-mini | OpenAI | closed | Estimated reference | 82% |
| #88 | Mistral Medium 3.5 128B mistral-medium-3-5-128b | Mistral AI | open | Estimated reference | 82% |
| #89 | GPT-5 (high) gpt-5-high | OpenAI | closed | Estimated reference | 82.1% |
| #90 | Nemotron 3 Nano 30B nemotron-3-nano-30b | NVIDIA | open | Estimated reference | 82.9% |
| #91 | Nemotron 3 Nano Omni 30B A3B nemotron-3-nano-omni-30b-a3b | NVIDIA | open | Estimated reference | 83.1% |
| #92 | Granite-4.0-H-1B granite-4-0-h-1b | IBM | open | Estimated reference | 83.4% |
| #93 | DeepSeek V3.1 deepseek-v3-1 | DeepSeek | open | Reference only | 83.5% |
| #94 | Mistral Large 3 mistral-large-2512 | Mistral AI | open | Reference only | 83.7% |
| #95 | DeepSeek-R1 deepseek-r1 | DeepSeek | open | Reference only | 84% |
| #96 | Qwen3.5-35B-A3B qwen3-5-35b-a3b | Alibaba Cloud | open | Reference only | 84% |
| #97 | Step 3.7 Flash step-3-7-flash | StepFun | open | Estimated reference | 84.4% |
| #98 | GPT-5.6 Terra gpt-5.6-terra | OpenAI | closed | Estimated reference | 85.2% |
| #99 | GPT-5.5 gpt-5.5 | OpenAI | closed | Estimated reference | 85.5% |
| #100 | Qwen3.5-122B-A10B qwen3-5-122b-a10b | Alibaba Cloud | open | Reference only | 85.5% |
| #101 | Trinity-Large-Preview trinity-large-preview | Arcee AI | open | Estimated reference | 86.6% |
| #102 | Trinity-Large-Thinking trinity-large-thinking | Arcee AI | open | Reference only | 86.6% |
| #103 | GPT-5.3-Codex gpt-5.3-codex | OpenAI | closed | Reference only | 86.9% |
| #104 | o3 o3 | OpenAI | closed | Reference only | 87.1% |
| #105 | Llama 4 Maverick llama-4-maverick | Meta | open | Reference only | 87.3% |
| #106 | Gemini 2.5 Pro gemini-2.5-pro | closed | Reference only | 87.4% | |
| #107 | DeepSeek V4 Pro (High) deepseek-v4-pro-high | DeepSeek | open | Estimated reference | 88.6% |
| #108 | GPT-5.4 gpt-5.4 | OpenAI | closed | Reference only | 88.6% |
| #109 | GPT-5.6 Sol gpt-5.6-sol | OpenAI | closed | Reference only | 88.8% |
| #110 | K-Exaone k-exaone | LG AI Research | closed | Estimated reference | 89.1% |
| #111 | Qwen3.5 397B qwen3.5-397b-a17b | Alibaba Cloud | open | Estimated reference | 89.1% |
| #112 | Qwen3.5 397B (Reasoning) qwen3-5-397b-reasoning | Alibaba Cloud | open | Estimated reference | 89.1% |
| #113 | DeepSeek V3 deepseek-v3 | DeepSeek | open | Reference only | 89.4% |
| #114 | Qwen3 Max qwen3-max | Alibaba Cloud | closed | Estimated reference | 89.4% |
| #115 | Gemma 3 27B gemma-3-27b | open | Reference only | 89.5% | |
| #116 | DeepSeek V4 Flash (High) deepseek-v4-flash-high | DeepSeek | open | Estimated reference | 89.7% |
| #117 | GPT-5.4 mini gpt-5.4-mini | OpenAI | closed | Estimated reference | 89.8% |
| #118 | GPT-5.6 Luna gpt-5.6-luna | OpenAI | closed | Estimated reference | 90.1% |
| #119 | Gemini 3 Flash gemini-3-flash | closed | Reference only | 90.2% | |
| #120 | GLM-4.7 glm-4.7 | Z.AI | open | Reference only | 90.3% |
| #121 | Gemini 3 Pro gemini-3-pro | closed | Reference only | 90.9% | |
| #122 | GPT-OSS 120B gpt-oss-120b | OpenAI | open | Reference only | 91.2% |
| #123 | Exaone 4.0 1.2B exaone-4-0-1-2b | LG AI Research | open | Estimated reference | 91.5% |
| #124 | Solar Pro 2 solar-pro-2 | Upstage | closed | Estimated reference | 91.5% |
| #125 | GLM-4.5-Air glm-4-5-air | Z.AI | closed | Reference only | 92.3% |
| #126 | Gemini 2.5 Flash gemini-2.5-flash | closed | Reference only | 93.3% | |
| #127 | DeepSeek V3.2 deepseek-v3.2 | DeepSeek | open | Reference only | 93.5% |
| #128 | Granite-4.0-1B granite-4-0-1b | IBM | open | Estimated reference | 93.5% |
| #129 | Sarvam 105B sarvam-105b | Sarvam | open | Estimated reference | 93.5% |
| #130 | DeepSeek V4 Pro (Max) deepseek-v4-pro-max | DeepSeek | open | Estimated reference | 94% |
| #131 | LFM2.5-VL-1.6B-Extract lfm2-5-vl-1-6b-extract | LiquidAI | open | Estimated reference | 94% |
| #132 | GPT-OSS 20B gpt-oss-20b | OpenAI | open | Reference only | 94.1% |
| #133 | Granite-4.0-H-350M granite-4-0-h-350m | IBM | open | Estimated reference | 94.4% |
| #134 | DeepSeek V4 Flash (Max) deepseek-v4-flash-max | DeepSeek | open | Estimated reference | 95.8% |
| #135 | Ling 2.6 Flash ling-2-6-flash | InclusionAI | open | Estimated reference | 95.8% |
| #136 | Sarvam 30B sarvam-30b | Sarvam | open | Estimated reference | 97% |
The top of this snapshot is led by Command A+ at 14.1%; third place is 8.8 points behind. The top-10 spread is 15.8 points.
About Artificial Analysis Omniscience Hallucination Rate
A display-only Artificial Analysis factuality metric for the rate of incorrect answers among non-correct responses. Results stay tied to the exact model variant and evaluation system. Multiple systems for the same model use the best published score on this page; overall Lumina scoring uses the median of ranking-eligible rows.
Open benchmark source ↗FAQ
A display-only Artificial Analysis factuality metric for the rate of incorrect answers among non-correct responses.
Command A+ by Cohere currently leads with 14.1%.
136 models in the LuminaBench cohort have a qualifying score on this benchmark.
Related
2026 · 21 results · Reference
Artificial Analysis MMLU-Pro2026 · 8 results · Reference
Massive Multitask Language Understanding2020 · 16 results · Reference
Graduate-Level Google-Proof Q&A2023 · 140 results · Reference
SuperGPQA: Scaling LLM Evaluation Across 285 Graduate Disciplines2025 · 38 results · Reference
Massive Multitask Language Understanding Professional2024 · 85 results · Reference
