Humanity's Last Exam (Artificial Analysis run) leaderboard
- Claude Opus 5.5 leads
- 61.4%
- models measured
- 192
- lead over #2
- +2.3 pts
- median
- 22.8%
Leader on Humanity's Last Exam (Artificial Analysis run) · read October 21
On October 21, 2026, 192 of our catalog's models had a Humanity's Last Exam (Artificial Analysis run) result, and Anthropic Claude Opus 5.5 led with 61.4%.
Expert-written questions across dozens of subjects. Share answered correctly. Run by Artificial Analysis on its own harness.
Role in the index
We collect this benchmark and show it as supporting evidence. It is not part of the Vibe Coding Index, because it measures something other than coding and the index is about coding. Read the methodology
| # | Model | Result | Evidence | Coding index | Output $/Mtok |
|---|---|---|---|---|---|
| 1 | Claude Opus 5.5Anthropic | 61.4% | 92.2 | $20.00 | |
| 2 | Claude Fable 5.1Anthropic | 59.1% | 89.1 | $50.00 | |
| 3 | Gemini 4 ArgonGoogle | 57.1% | 80.5 | $10.00 | |
| 4 | Claude Fable 5Anthropic | 55.5% | 82.1 | $50.00 | |
| 5 | Claude Sonnet 5.5Anthropic | 55.0% | 88.9 | $10.00 | |
| 6 | Claude Opus 5Anthropic | 54.9% | 81.8 | $25.00 | |
| 7 | GPT-6 AstraOpenAI | 54.7% | 88.7 | $50.00 | |
| 8 | GPT-6.1 SolOpenAI | 52.9% | 88.7 | $10.00 | |
| 9 | Muse WonderMeta | 50.1% | 64.5 | $20.00 | |
| 10 | GPT-5.6 SolOpenAI | 49.5% | 76.7 | $20.00 | |
| 11 | MiMo-V2.6-ProXiaomi | 49.4% | 72.6 | $0.87 | |
| 12 | Claude Opus 4.8Anthropic | 48.7% | 66.0 | $25.00 | |
| 13 | Muse Spark 1.3Meta | 48.7% | 79.7 | $4.25 | |
| 14 | Gemini 3.7 FlashGoogle | 47.9% | 69.9 | $3.75 | |
| 15 | GPT-6 SolOpenAI | 47.9% | 82.8 | $10.00 | |
| 16 | Gemini 3.8 FlashGoogle | 47.8% | 70.5 | $4.125 | |
| 17 | Gemini 3.1 Pro PreviewGoogle | 47.0% | 35.3 | $12.00 | |
| 18 | Kimi K3Moonshot AI | 46.9% | 61.8 | $11.936 | |
| 19 | Step 5 PreviewStepFun | 46.5% | 71.4 | $2.70 | |
| 20 | Muse Spark 1.1Meta | 46.2% | 48.9 | $4.25 | |
| 21 | GPT-5.5OpenAI | 45.8% | 67.8 | $30.00 | |
| 22 | Muse Spark 1.2Meta | 45.5% | 65.1 | $4.25 | |
| 23 | Claude Haiku 5.5Anthropic | 44.4% | 75.9 | $0.50 | |
| 24 | Grok 4.6xAI | 44.1% | 76.4 | $6.00 | |
| 25 | Muse Spark OpenMeta | 43.9% | 62.1 | $4.25 | |
| 26 | GPT-5.4OpenAI | 43.7% | 37.7 | $15.00 | |
| 27 | DeepSeek V4.1 Flash 1017DeepSeek | 43.3% | 83.3 | $0.60 | |
| 28 | Qwen3.8 Max (0902)Qwen | 43.1% | 63.5 | $6.00 | |
| 29 | Grok 4.7xAI | 43.1% | 75.0 | $6.00 | |
| 30 | GPT-5.6 TerraOpenAI | 42.9% | 69.0 | $12.00 | |
| 31 | Gemini 3.5 FlashGoogle | 42.7% | 42.8 | $9.00 | |
| 32 | Grok 4.5xAI | 42.7% | 75.5 | $6.00 | |
| 33 | GPT-5.3-CodexOpenAI | 42.5% | 29.5 | $14.00 | |
| 34 | Qwen3.8 2.4T A95BQwen | 42.4% | Not benchmarked | $6.00 | |
| 35 | Claude Opus 4.7Anthropic | 42.3% | 51.1 | $25.00 | |
| 36 | GLM 5.3Z.ai | 42.3% | 73.1 | $1.40 | |
| 37 | Claude Sonnet 5Anthropic | 41.3% | 63.5 | $10.00 | |
| 38 | GLM 5.2Z.ai | 41.1% | 57.7 | $7.00 | |
| 39 | DeepSeek V4 Pro 0813DeepSeek | 41.0% | 54.6 | $1.98 | |
| 40 | Gemini 3.6 FlashGoogle | 40.8% | 48.3 | $3.75 | |
| 41 | Qwen3.7 MaxQwen | 40.5% | 45.1 | $4.425 | |
| 42 | Claude Opus 4.6Anthropic | 39.9% | 66.9 | $25.00 | |
| 43 | GLM 5.3 FlashZ.ai | 39.9% | 71.0 | $0.50 | |
| 44 | GPT-5.6 LunaOpenAI | 39.5% | 61.5 | $1.20 | |
| 45 | Ling 3.1 FlashinclusionAI | 39.4% | Not benchmarked | $0.225 | |
| 46 | DeepSeek V4.1 FlashDeepSeek | 39.2% | 77.2 | $0.525 | |
| 47 | MiniMax M3MiniMax | 39.0% | 57.8 | $1.20 | |
| 48 | DeepSeek V4 Flash 0731DeepSeek | 38.6% | Not benchmarked | $1.32 | |
| 49 | GPT-6 LunaOpenAI | 38.5% | 71.4 | $0.50 | |
| 50 | GPT-5.2OpenAI | 37.7% | 58.9 | $14.00 | |
| 51 | Kimi K2.6Moonshot AI | 37.5% | 44.2 | $3.41 | |
| 52 | Grok 4.3xAI | 37.2% | 22.5 | $2.50 | |
| 53 | Gemini 3 Flash PreviewGoogle | 36.6% | 75.8 | $3.00 | |
| 54 | GPT-5.2-CodexOpenAI | 35.7% | 55.1 | $14.00 | |
| 55 | MiMo-V2.5-ProXiaomi | 35.7% | 40.0 | $0.609 | |
| 56 | Qwen3.7 PlusQwen | 35.6% | Not benchmarked | $1.28 | |
| 57 | MiMo-V2.6-FlashXiaomi | 35.1% | 61.9 | $0.28 | |
| 58 | Mistral Large 4Mistral AI | 35.0% | 55.3 | $4.18 | |
| 59 | Kimi K2.7 CodeMoonshot AI | 35.0% | 39.1 | $3.35 | |
| 60 | DeepSeek V4 Flash Vision ExpDeepSeek | 34.5% | 37.6 | $0.6468 | |
| 61 | Grok 4.20xAI | 34.5% | Not benchmarked | $2.50 | |
| 62 | Qwen3.8 27BQwen | 33.9% | 73.6 | $1.35 | |
| 63 | LongCat 2.0Meituan | 33.7% | Not benchmarked | $1.20 | |
| 64 | Claude Sonnet 4.6Anthropic | 33.6% | 57.8 | $15.00 | |
| 65 | Hy3Tencent | 33.5% | 43.9 | $0.528 | |
| 66 | T Inkling SmallThinking Machines | 33.3% | 29.9 | $1.20 | |
| 67 | T InklingThinking Machines | 31.9% | 49.7 | $4.05 | |
| 68 | Kimi K2.5Moonshot AI | 30.7% | 58.5 | $2.50 | |
| 69 | Nemotron 3.5 Super VLNVIDIA | 30.7% | 65.5 | $0.50 | |
| 70 | Claude Opus 4.5Anthropic | 30.1% | 66.8 | $25.00 | |
| 71 | GLM 5.1Z.ai | 30.1% | 44.0 | $4.40 | |
| 72 | Solar Pro 4Upstage | 29.8% | 24.4 | $0.60 | |
| 73 | MiniMax M2.7MiniMax | 29.6% | 28.4 | $1.20 | |
| 74 | GLM 5Z.ai | 29.3% | 59.7 | $3.20 | |
| 75 | Qwen3.5 397B A17BQwen | 29.0% | 28.6 | $2.34 | |
| 76 | GPT-5OpenAI | 28.5% | 72.2 | $10.00 | |
| 77 | GPT-5.1OpenAI | 28.5% | 53.3 | $10.00 | |
| 78 | Nemotron 3 UltraNVIDIA | 28.4% | Not benchmarked | $2.40 | |
| 79 | GPT-5.4 NanoOpenAI | 28.3% | Not benchmarked | $1.25 | |
| 80 | GPT-5.4 MiniOpenAI | 28.1% | 28.2 | $4.50 | |
| 81 | Qwen3.6 PlusQwen | 27.8% | 37.4 | $1.95 | |
| 82 | Hy3 previewTencent | 27.8% | 22.4 | $0.60 | |
| 83 | GLM 5 TurboZ.ai | 27.8% | Not benchmarked | $4.00 | |
| 84 | GLM 4.7Z.ai | 27.4% | 33.6 | $1.98 | |
| 85 | MiMo-V2.5Xiaomi | 27.2% | 34.0 | $0.238 | |
| 86 | Solar Mini 4Upstage | 25.8% | Not benchmarked | $0.20 | |
| 87 | GPT-5.1-CodexOpenAI | 25.7% | 50.5 | $10.00 | |
| 88 | Qwen3.5-122B-A10BQwen | 25.2% | 22.9 | $2.08 | |
| 89 | DeepSeek V3.2DeepSeek | 24.6% | 54.4 | $0.42 | |
| 90 | Step 3.5 FlashStepFun | 24.5% | Not benchmarked | $0.30 | |
| 91 | Qwen3.5-27BQwen | 23.9% | 22.6 | $2.16 | |
| 92 | Kimi K2 ThinkingMoonshot AI | 23.8% | 63.4 | $2.50 | |
| 93 | Ling 3.0 FlashinclusionAI | 23.7% | Not benchmarked | $0.063 | |
| 94 | Gemma 4 31BGoogle | 23.6% | 38.3 | $0.40 | |
| 95 | MiniMax M2.1MiniMax | 23.2% | Not benchmarked | $1.20 | |
| 96 | Qwen3.6 27BQwen | 23.1% | Not benchmarked | $2.00 | |
| 97 | Gemini 2.5 ProGoogle | 22.5% | 37.0 | $10.00 | |
| 98 | Qwen3.6 35B A3BQwen | 22.2% | Not benchmarked | $0.95 | |
| 99 | Ling 3.0 Flash VLinclusionAI | 22.0% | Not benchmarked | $0.18 | |
| 100 | GPT-5 MiniOpenAI | 21.5% | 59.8 | $2.00 | |
| 101 | Step 3.7 FlashStepFun | 21.4% | Not benchmarked | $1.15 | |
| 102 | Qwen3.5-35B-A3BQwen | 21.0% | 7.7 | $1.00 | |
| 103 | Nemotron 3 SuperNVIDIA | 20.8% | Not benchmarked | $2.40 | |
| 104 | MiniMax M2.5MiniMax | 20.5% | 59.4 | $0.95 | |
| 105 | o3OpenAI | 20.1% | 70.7 | $8.00 | |
| 106 | Gemini 2.5 Pro Preview 06-05Google | 19.9% | 83.1 | $10.00 | |
| 107 | gpt-oss-120bOpenAI | 19.6% | 34.5 | $0.60 | |
| 108 | Gemma 4 26B A4BGoogle | 19.3% | 22.8 | $0.255 | |
| 109 | Gemini 3.5 Flash LiteGoogle | 18.8% | 33.6 | $2.50 | |
| 110 | GPT-5.1-Codex-MiniOpenAI | 18.5% | 6.2 | $2.00 | |
| 111 | Gemini 3.1 Flash LiteGoogle | 17.2% | Not benchmarked | $1.50 | |
| 112 | Mercury 2Inception | 17.1% | 0.0 | $0.75 | |
| 113 | GLM 5V TurboZ.ai | 17.1% | 28.8 | $4.00 | |
| 114 | o4 MiniOpenAI | 16.5% | 68.6 | $4.40 | |
| 115 | DeepSeek V3.1 TerminusDeepSeek | 16.4% | Not benchmarked | $1.00 | |
| 116 | Qwen3 235B A22B Instruct 2507Qwen | 15.9% | Not benchmarked | $0.55 | |
| 117 | Trinity Large ThinkingArcee AI | 15.8% | 5.4 | $0.80 | |
| 118 | R1 0528DeepSeek | 15.8% | 71.4 | $2.15 | |
| 119 | DeepSeek V3.2 ExpDeepSeek | 14.9% | 55.0 | $0.41 | |
| 120 | Qwen3.5-9BQwen | 14.9% | Not benchmarked | $0.13 | |
| 121 | GLM 4.6Z.ai | 14.5% | 52.2 | $2.20 | |
| 122 | Mistral Medium 3.5Mistral AI | 13.8% | 9.2 | $7.50 | |
| 123 | MiniMax M2MiniMax | 13.7% | 45.3 | $1.02 | |
| 124 | GLM 4.5Z.ai | 13.0% | 64.2 | $2.20 | |
| 125 | Qwen3 Next 80B A3B InstructQwen | 12.6% | Not benchmarked | $1.10 | |
| 126 | Gemini 2.5 FlashGoogle | 12.1% | 28.7 | $2.50 | |
| 127 | Command ACohere | 12.0% | Not benchmarked | $10.00 | |
| 128 | o3 MiniOpenAI | 12.0% | 52.1 | $4.40 | |
| 129 | Qwen3 VL 235B A22B InstructQwen | 11.9% | Not benchmarked | $1.90 | |
| 130 | Mercury 2.5Inception | 11.8% | Not benchmarked | $0.15 | |
| 131 | Nemotron 3 Nano 30B A3BNVIDIA | 11.4% | Not benchmarked | $0.24 | |
| 132 | gpt-oss-20bOpenAI | 11.0% | Not benchmarked | $0.09 | |
| 133 | Nemotron 3.5 LightningNVIDIA | 10.6% | Not benchmarked | $0.20 | |
| 134 | Qwen3 30B A3B Instruct 2507Qwen | 10.3% | Not benchmarked | $0.30 | |
| 135 | Solar Pro 3Upstage | 10.3% | Not benchmarked | $0.60 | |
| 136 | Qwen3 Coder NextQwen | 10.1% | Not benchmarked | $0.80 | |
| 137 | Mistral Small 4Mistral AI | 9.9% | Not benchmarked | $0.60 | |
| 138 | Granite 4.2 8BIBM Granite | 9.7% | Not benchmarked | $0.25 | |
| 139 | GLM 4.6VZ.ai | 9.6% | Not benchmarked | $0.90 | |
| 140 | GPT-5 NanoOpenAI | 9.5% | 34.8 | $0.40 | |
| 141 | Qwen3 VL 30B A3B InstructQwen | 8.9% | Not benchmarked | $0.60 | |
| 142 | GLM 4.7 FlashZ.ai | 7.6% | Not benchmarked | $0.40 | |
| 143 | Kimi K2 0711Moonshot AI | 7.4% | 64.7 | $2.30 | |
| 144 | Qwen3 32BQwen | 7.4% | 40.0 | $0.28 | |
| 145 | o1OpenAI | 7.0% | 61.7 | $60.00 | |
| 146 | GLM 4.5 AirZ.ai | 7.0% | Not benchmarked | $0.85 | |
| 147 | Gemini 2.5 Flash LiteGoogle | 6.8% | Not benchmarked | $0.40 | |
| 148 | Sonar ProPerplexity | 6.6% | Not benchmarked | $15.00 | |
| 149 | Kimi K2 0905Moonshot AI | 6.4% | Not benchmarked | $2.50 | |
| 150 | GLM 4.5VZ.ai | 6.3% | Not benchmarked | $1.80 | |
| 151 | Qwen3 30B A3BQwen | 6.2% | Not benchmarked | $0.50 | |
| 152 | Llama 3.2 1B InstructMeta Llama | 5.5% | Not benchmarked | $0.201 | |
| 153 | Gemma 3 4BGoogle | 5.3% | Not benchmarked | $0.05 | |
| 154 | Llama 3.1 8B InstructMeta Llama | 5.3% | Not benchmarked | $0.08 | |
| 155 | Llama 3.2 3B InstructMeta Llama | 5.3% | Not benchmarked | $0.335 | |
| 156 | Granite 4.0 MicroIBM Granite | 5.0% | Not benchmarked | $0.112 | |
| 157 | GPT-4.1 MiniOpenAI | 5.0% | 28.5 | $1.60 | |
| 158 | Llama 4 MaverickMeta Llama | 4.9% | 18.1 | $0.6525 | |
| 159 | SonarPerplexity | 4.9% | Not benchmarked | $1.00 | |
| 160 | DeepSeek V3 0324DeepSeek | 4.7% | 42.0 | $1.14 | |
| 161 | Mistral Medium 3.1Mistral AI | 4.7% | Not benchmarked | $2.00 | |
| 162 | Nova Micro 1.0Amazon | 4.6% | Not benchmarked | $0.14 | |
| 163 | Llama 3.1 70B InstructMeta Llama | 4.5% | Not benchmarked | $0.40 | |
| 164 | Qwen3 14BQwen | 4.5% | Not benchmarked | $0.24 | |
| 165 | Qwen3 Coder 480B A35BQwen | 4.5% | 50.0 | $1.00 | |
| 166 | Gemma 3 27BGoogle | 4.4% | 4.9 | $0.35 | |
| 167 | Reka Flash 3Reka | 4.4% | Not benchmarked | $0.20 | |
| 168 | Nova Lite 1.0Amazon | 4.3% | Not benchmarked | $0.24 | |
| 169 | SabaMistral AI | 4.3% | Not benchmarked | $0.60 | |
| 170 | Nova Premier 1.0Amazon | 4.2% | Not benchmarked | $12.50 | |
| 171 | Gemma 3 12BGoogle | 4.2% | Not benchmarked | $0.15 | |
| 172 | GPT-4.1OpenAI | 4.2% | 46.5 | $8.00 | |
| 173 | GPT-4o-miniOpenAI | 4.2% | Not benchmarked | $0.60 | |
| 174 | Mistral Medium 3Mistral AI | 4.1% | Not benchmarked | $2.00 | |
| 175 | Mixtral 8x22B InstructMistral AI | 4.0% | Not benchmarked | $6.00 | |
| 176 | N Hermes 3 70B InstructNous Research | 4.0% | Not benchmarked | $0.70 | |
| 177 | Llama 4 ScoutMeta Llama | 3.8% | 9.1 | $0.30 | |
| 178 | Phi 4Microsoft | 3.8% | Not benchmarked | $0.14 | |
| 179 | Mistral Small 3Mistral AI | 3.8% | Not benchmarked | $0.08 | |
| 180 | GPT-4.1 NanoOpenAI | 3.8% | 8.9 | $0.40 | |
| 181 | Qwen3 Coder 30B A3B InstructQwen | 3.8% | 60.4 | $0.28 | |
| 182 | Qwen3 VL 8B InstructQwen | 3.8% | Not benchmarked | $0.75 | |
| 183 | Llama 3.3 70B InstructMeta Llama | 3.6% | Not benchmarked | $0.32 | |
| 184 | Qwen2.5 72B InstructQwen | 3.6% | Not benchmarked | $0.40 | |
| 185 | Mistral LargeMistral AI | 3.5% | Not benchmarked | $6.00 | |
| 186 | Qwen2.5 Coder 32B InstructQwen | 3.5% | 30.5 | $1.00 | |
| 187 | Nova Pro 1.0Amazon | 3.2% | Not benchmarked | $3.20 | |
| 188 | GPT-4 TurboOpenAI | 3.1% | Not benchmarked | $30.00 | |
| 189 | Mistral Large 2407Mistral AI | 2.9% | Not benchmarked | $6.00 | |
| 190 | GPT-4oOpenAI | 2.4% | 38.8 | $10.00 | |
| 191 | GPT-4o (2024-08-06)OpenAI | 2.3% | 23.1 | $10.00 | |
| 192 | GPT-4o (2024-05-13)OpenAI | 1.8% | Not benchmarked | $15.00 |
Only models in our catalog are listed, and each appears once with its best published run. 195 rows were read from the source on October 21, 2026, and rows that match no catalog model are left out.