| # | Model | Score | Relative | Setting | Released | Run | Evidence |
|---|---|---|---|---|---|---|---|
| 1 | Claude Fable 5 Anthropic | 98.4% | max | 9 Jun 2026 | — | Source ↗ | |
| 2 | Claude Opus 5 Anthropic | 97.9% | high | 24 Jul 2026 | — | Source ↗ | |
| 3 | Claude Opus 5 (xhigh) Anthropic | 97.9% | xhigh | 24 Jul 2026 | — | Source ↗ | |
| 4 | Claude Fable 5.1 (xhigh) Anthropic | 97.6% | xhigh | 1 Sep 2026 | — | Source ↗ | |
| 5 | Claude Fable 5.1 (unknown) Anthropic | 97.6% | unknown | 1 Sep 2026 | — | Source ↗ | |
| 6 | GPT-6 Astra (xhigh) OpenAI | 97.3% | xhigh | 3 Sep 2026 | — | Source ↗ | |
| 7 | Grok 4.6 (xhigh) xAI | 97.3% | xhigh | 12 Aug 2026 | — | Source ↗ | |
| 8 | Claude Fable 5 (xhigh) Anthropic | 97.3% | xhigh | 9 Jun 2026 | — | Source ↗ | |
| 9 | GPT-6 Astra (medium) OpenAI | 97.1% | medium | 3 Sep 2026 | — | Source ↗ | |
| 10 | GPT-6 Astra (unknown thinking) OpenAI | 97.1% | unknown | 3 Sep 2026 | — | Source ↗ | |
| 11 | Claude Fable 5.1 (medium) Anthropic | 97.1% | medium | 1 Sep 2026 | — | Source ↗ | |
| 12 | Gemini 3.1 Pro Preview Google | 97.1% | medium | 19 Feb 2026 | — | Source ↗ | |
| 13 | Gemini 3.7 Flash Google | 96.8% | high | 13 Aug 2026 | — | Source ↗ | |
| 14 | GPT-6 Astra OpenAI | 96.5% | high | 3 Sep 2026 | — | Source ↗ | |
| 15 | Muse Spark 1.3 (unknown) Meta AI | 96.5% | unknown | 2 Sep 2026 | — | Source ↗ | |
| 16 | Claude Fable 5.1 Anthropic | 96.5% | high | 1 Sep 2026 | — | Source ↗ | |
| 17 | Grok 4.5 SpaceXAI | 96.5% | high | 8 Jul 2026 | — | Source ↗ | |
| 18 | GPT-5.6 Sol (pro, max) OpenAI | 96.0% | promax | 9 Jul 2026 | — | Source ↗ | |
| 19 | GPT-5.5 Pro (xhigh) OpenAI | 96.0% | xhigh | 23 Apr 2026 | — | Source ↗ | |
| 20 | GPT-5.5 (xhigh) OpenAI | 96.0% | xhigh | 23 Apr 2026 | — | Source ↗ | |
| 21 | GPT-5.6 Sol (xhigh) OpenAI | 95.7% | xhigh | 9 Jul 2026 | — | Source ↗ | |
| 22 | Claude Fable 5 (medium) Anthropic | 95.7% | medium | 9 Jun 2026 | — | Source ↗ | |
| 23 | Gemini 3.8 Flash (unknown) Google DeepMind | 95.7% | unknown | 2 Sep 2026 | — | Source ↗ | |
| 24 | Gemini 3.6 Flash Google | 95.5% | high | 21 Jul 2026 | — | Source ↗ | |
| 25 | GPT-5.6 Sol OpenAI | 95.5% | max | 9 Jul 2026 | — | Source ↗ | |
| 26 | GPT-5.6 Sol (medium) OpenAI | 95.2% | medium | 9 Jul 2026 | — | Source ↗ | |
| 27 | GPT-6 Astra (low) OpenAI | 94.9% | low | 3 Sep 2026 | — | Source ↗ | |
| 28 | Gemini 3.8 Flash (medium) Google DeepMind | 94.9% | medium | 2 Sep 2026 | — | Source ↗ | |
| 29 | Gemini 3.8 Flash (low) Google DeepMind | 94.9% | low | 2 Sep 2026 | — | Source ↗ | |
| 30 | Claude Opus 5 (medium) Anthropic | 94.9% | medium | 24 Jul 2026 | — | Source ↗ | |
| 31 | Claude Opus 4.8 Anthropic | 94.9% | max | 28 May 2026 | — | Source ↗ | |
| 32 | Gemini 3.7 Flash (medium) Google DeepMind | 94.7% | medium | 13 Aug 2026 | — | Source ↗ | |
| 33 | Muse Spark 1.2 (xhigh) Meta AI | 94.7% | xhigh | 5 Aug 2026 | — | Source ↗ | |
| 34 | Gemini 3.5 Flash Google | 94.7% | high | 19 May 2026 | — | Source ↗ | |
| 35 | Claude Opus 4.7 Anthropic | 94.7% | max | 16 Apr 2026 | — | Source ↗ | |
| 36 | Muse Spark 1.1 Meta | 94.4% | high | 9 Jul 2026 | — | Source ↗ | |
| 37 | GPT-5.4 (xhigh) OpenAI | 94.4% | xhigh | 5 Mar 2026 | — | Source ↗ | |
| 38 | Claude Fable 5.1 (low) Anthropic | 94.1% | low | 1 Sep 2026 | — | Source ↗ | |
| 39 | DeepSeek V4 Pro 0813 (unknown) Open source DeepSeek | 93.9% | unknown | 13 Aug 2026 | — | Source ↗ | |
| 40 | GPT-5.6 Sol (low) OpenAI | 93.9% | low | 9 Jul 2026 | — | Source ↗ | |
| 41 | GLM 5.2 Open source Z.ai | 93.6% | max | 16 Jun 2026 | — | Source ↗ | |
| 42 | Gemini 3.7 Flash (low) Google DeepMind | 93.3% | low | 13 Aug 2026 | — | Source ↗ | |
| 43 | GPT-5.6 Terra OpenAI | 93.3% | max | 9 Jul 2026 | — | Source ↗ | |
| 44 | Claude Opus 5 (low) Anthropic | 93.1% | low | 24 Jul 2026 | — | Source ↗ | |
| 45 | GPT-5.6 Terra (xhigh) OpenAI | 92.5% | xhigh | 9 Jul 2026 | — | Source ↗ | |
| 46 | Claude Sonnet 5 Anthropic | 92.5% | max | 30 Jun 2026 | — | Source ↗ | |
| 47 | Qwen3.7 Max Qwen | 92.3% | — | 19 May 2026 | — | Source ↗ | |
| 48 | Qwen3.8 Max (xhigh) Alibaba | 92.0% | xhigh | 2 Aug 2026 | — | Source ↗ | |
| 49 | Claude Fable 5 (low) Anthropic | 91.5% | low | 9 Jun 2026 | — | Source ↗ | |
| 50 | Kimi K3 Open source MoonshotAI | 91.2% | max | 16 Jul 2026 | — | Source ↗ | |
| 51 | Claude Opus 4.6 Anthropic | 91.2% | max | 5 Feb 2026 | — | Source ↗ | |
| 52 | DeepSeek V4 Flash 0731 Open source DeepSeek | 90.9% | max | 31 Jul 2026 | — | Source ↗ | |
| 53 | Kimi K2.6 Open source MoonshotAI | 90.9% | — | 20 Apr 2026 | — | Source ↗ | |
| 54 | GPT-5.2 (xhigh) OpenAI | 90.9% | xhigh | 11 Dec 2025 | — | Source ↗ | |
| 55 | DeepSeek v4 Pro (high) Open source DeepSeek | 90.7% | max | 24 Apr 2026 | — | Source ↗ | |
| 56 | Grok 4.3 Beta xAI | 90.7% | high | 17 Apr 2026 | — | Source ↗ | |
| 57 | GPT-5 OpenAI | 90.7% | high | 7 Aug 2025 | — | Source ↗ | |
| 58 | GPT-5.6 Terra (low) OpenAI | 90.1% | low | 9 Jul 2026 | — | Source ↗ | |
| 59 | GPT-5.6 Terra (medium) OpenAI | 90.1% | medium | 9 Jul 2026 | — | Source ↗ | |
| 60 | Nemotron 3 Ultra Open source NVIDIA | 90.1% | — | 4 Jun 2026 | — | Source ↗ | |
| 61 | Grok 4.20 SpaceXAI | 90.1% | — | 17 Feb 2026 | — | Source ↗ | |
| 62 | GPT-5.1 OpenAI | 90.1% | high | 13 Nov 2025 | — | Source ↗ | |
| 63 | DeepSeek V4.1 Flash (unknown) Open source DeepSeek | 89.9% | unknown | 9 Sep 2026 | — | Source ↗ | |
| 64 | Claude Sonnet 4.6 Anthropic | 89.9% | max | 17 Feb 2026 | — | Source ↗ | |
| 65 | Claude Opus 4.5 Anthropic | 89.9% | high | 24 Nov 2025 | — | Source ↗ | |
| 66 | Claude Sonnet 5 (xhigh) Anthropic | 89.5% | xhigh | 30 Jun 2026 | — | Source ↗ | |
| 67 | GPT-5.6 Luna (xhigh) OpenAI | 89.1% | xhigh | 9 Jul 2026 | — | Source ↗ | |
| 68 | Gemini 3 Flash Preview Google | 89.1% | high | 17 Dec 2025 | — | Source ↗ | |
| 69 | GPT-5.6 Luna OpenAI | 88.8% | max | 9 Jul 2026 | — | Source ↗ | |
| 70 | Qwen3.8 27B Open source Qwen | 88.0% | — | 14 Aug 2026 | — | Source ↗ | |
| 71 | Grok 4.1 Fast xAI | 87.7% | — | 19 Nov 2025 | — | Source ↗ | |
| 72 | DeepSeek V3.2 Exp Open source DeepSeek | 87.7% | thinking | 29 Sep 2025 | — | Source ↗ | |
| 73 | GLM-5.3 (unknown thinking) Z.ai (Zhipu AI) | 87.7% | unknown | 14 Aug 2026 | — | Source ↗ | |
| 74 | Inkling (xhigh) Open source Thinking Machines | 87.5% | xhigh | 15 Jul 2026 | — | Source ↗ | |
| 75 | Qwen3.5 397B A17B Open source Qwen | 87.5% | — | 13 Feb 2026 | — | Source ↗ | |
| 76 | Qwen 3.6 Max (Preview) Alibaba | 87.2% | — | 20 Apr 2026 | — | Source ↗ | |
| 77 | o3 Pro OpenAI | 86.9% | high | 10 Jun 2025 | — | Source ↗ | |
| 78 | DeepSeek v4 Flash (max) Open source DeepSeek | 86.4% | max | 24 Apr 2026 | — | Source ↗ | |
| 79 | DeepSeek-V3.2 (Thinking; Novita) Open source DeepSeek | 85.6% | — | 1 Dec 2025 | — | Source ↗ | |
| 80 | o3 OpenAI | 84.8% | high | 16 Apr 2025 | — | Source ↗ | |
| 81 | MiMo-V2.5-Pro Open source Xiaomi | 84.5% | — | 22 Apr 2026 | — | Source ↗ | |
| 82 | Qwen3.5-122B-A10B Open source Qwen | 84.3% | none | 25 Feb 2026 | — | Source ↗ | |
| 83 | Qwen3.7 Plus Qwen | 84.0% | — | 2 Jun 2026 | — | Source ↗ | |
| 84 | GPT-5.6 Sol (none) OpenAI | 83.7% | none | 9 Jul 2026 | — | Source ↗ | |
| 85 | Gemini 3.5 Flash Lite Google | 83.5% | high | 21 Jul 2026 | — | Source ↗ | |
| 86 | GPT-5.6 Luna (medium) OpenAI | 83.2% | medium | 9 Jul 2026 | — | Source ↗ | |
| 87 | Claude Sonnet 4.5 (unknown thinking) Anthropic | 83.2% | unknown | 29 Sep 2025 | — | Source ↗ | |
| 88 | Qwen3.5-Flash Qwen | 82.9% | — | 25 Feb 2026 | — | Source ↗ | |
| 89 | Gemma 4 31B Open source Google | 82.7% | — | 2 Apr 2026 | — | Source ↗ | |
| 90 | Grok 4 Fast xAI | 82.7% | — | 19 Sep 2025 | — | Source ↗ | |
| 91 | DeepSeek V3.1 Open source DeepSeek | 82.7% | thinking | 21 Aug 2025 | — | Source ↗ | |
| 92 | Qwen3.5-27B Open source Qwen | 82.4% | — | 25 Feb 2026 | — | Source ↗ | |
| 93 | Gemini 2.5 Pro (Jun 2025) Google DeepMind | 82.4% | — | 5 Jun 2025 | — | Source ↗ | |
| 94 | Qwen3-Max-Instruct Alibaba | 82.1% | — | 24 Sep 2025 | — | Source ↗ | |
| 95 | Qwen 3.6 Plus (2026-04-02) Alibaba | 81.9% | — | 31 Mar 2026 | — | Source ↗ | |
| 96 | Claude Opus 4 Anthropic | 81.6% | unknown | 22 May 2025 | — | Source ↗ | |
| 97 | DeepSeek V3.1 Terminus Open source DeepSeek | 81.3% | — | 22 Sep 2025 | — | Source ↗ | |
| 98 | Qwen Plus (Apr 2025) Alibaba | 81.1% | — | 28 Apr 2025 | — | Source ↗ | |
| 99 | Qwen3.5 Plus 2026-02-15 Qwen | 80.5% | — | 16 Feb 2026 | — | Source ↗ | |
| 100 | GPT-5 Mini OpenAI | 80.5% | high | 7 Aug 2025 | — | Source ↗ | |
| 101 | GPT-5.4 nano (xhigh) OpenAI | 80.3% | xhigh | 17 Mar 2026 | — | Source ↗ | |
| 102 | Qwen3-235B-A22B-Thinking (Jul 2025) Open source Alibaba | 80.3% | — | 25 Jul 2025 | — | Source ↗ | |
| 103 | GPT-5.6 Luna (low) OpenAI | 80.0% | low | 9 Jul 2026 | — | Source ↗ | |
| 104 | GPT-5.4 mini (xhigh) OpenAI | 80.0% | xhigh | 17 Mar 2026 | — | Source ↗ | |
| 105 | Qwen3.5-35B-A3B Open source Qwen | 80.0% | — | 24 Feb 2026 | — | Source ↗ | |
| 106 | Claude Opus 4.1 (unknown thinking) Anthropic | 80.0% | unknown | 5 Aug 2025 | — | Source ↗ | |
| 107 | Claude Sonnet 5 (medium) Anthropic | 79.2% | medium | 30 Jun 2026 | — | Source ↗ | |
| 108 | MiniMax M3 Open source MiniMax | 78.9% | — | 1 Jun 2026 | — | Source ↗ | |
| 109 | GPT-5.6 Terra (none) OpenAI | 78.7% | none | 9 Jul 2026 | — | Source ↗ | |
| 110 | Qwen3-235B-A22B-Instruct (Jul 2025) Open source Alibaba | 78.4% | — | 25 Jul 2025 | — | Source ↗ | |
| 111 | Qwen3.6 27B Open source Qwen | 78.1% | — | 22 Apr 2026 | — | Source ↗ | |
| 112 | o4 Mini OpenAI | 77.6% | high | 16 Apr 2025 | — | Source ↗ | |
| 113 | Qwen 3.6 Flash (2026-04-16) Alibaba | 77.1% | — | 27 Apr 2026 | — | Source ↗ | |
| 114 | Claude Sonnet 4 (unknown thinking) Anthropic | 77.1% | unknown | 22 May 2025 | — | Source ↗ | |
| 115 | Gemini 3.1 Flash Lite Google | 76.8% | high | 3 Mar 2026 | — | Source ↗ | |
| 116 | Gemini 2.5 Flash (Jun 2025) Google DeepMind | 76.5% | — | 17 Jun 2025 | — | Source ↗ | |
| 117 | gpt-oss-120b Open source OpenAI | 76.3% | high | 5 Aug 2025 | — | Source ↗ | |
| 118 | Qwen3 235B A22B Open source Qwen | 75.7% | — | 28 Apr 2025 | — | Source ↗ | |
| 119 | Mistral Medium 3.5 Open source Mistral | 75.5% | — | 28 Apr 2026 | — | Source ↗ | |
| 120 | Gemma 4 26B A4B Open source Google | 74.9% | — | 2 Apr 2026 | — | Source ↗ | |
| 121 | o1 OpenAI | 74.7% | high | 17 Dec 2024 | — | Source ↗ | |
| 122 | Qwen3.6 35B A3B Open source Qwen | 73.9% | — | 27 Apr 2026 | — | Source ↗ | |
| 123 | Claude Sonnet 5 (low) Anthropic | 73.6% | low | 30 Jun 2026 | — | Source ↗ | |
| 124 | Claude Haiku 4.5 (unknown thinking) Anthropic | 73.6% | unknown | 15 Oct 2025 | — | Source ↗ | |
| 125 | Qwen3.5-9B Open source Qwen | 71.2% | — | 24 Feb 2026 | — | Source ↗ | |
| 126 | Mistral Small 4 Open source Mistral | 70.9% | — | 16 Mar 2026 | — | Source ↗ | |
| 127 | Qwen3-30B-A3B-Thinking (Jul 2025) Open source Alibaba | 69.3% | — | 30 Jul 2025 | — | Source ↗ | |
| 128 | GPT-5.6 Luna (none) OpenAI | 69.3% | none | 9 Jul 2026 | — | Source ↗ | |
| 129 | GPT-4.1 Mini OpenAI | 68.8% | — | 14 Apr 2025 | — | Source ↗ | |
| 130 | o3 Mini OpenAI | 68.8% | high | 31 Jan 2025 | — | Source ↗ | |
| 131 | GPT-4.1 OpenAI | 68.3% | — | 14 Apr 2025 | — | Source ↗ | |
| 132 | gpt-oss-20b Open source OpenAI | 68.0% | high | 5 Aug 2025 | — | Source ↗ | |
| 133 | Claude 3.5 Sonnet (Oct 2024) Anthropic | 67.8% | — | 22 Oct 2024 | — | Source ↗ | |
| 134 | Claude 3.5 Sonnet (Jun 2024) Anthropic | 67.8% | — | 20 Jun 2024 | — | Source ↗ | |
| 135 | Mistral Medium 3.1 Mistral | 67.5% | — | 13 Aug 2025 | — | Source ↗ | |
| 136 | Qwen3 32B Open source Qwen | 67.5% | — | 29 Apr 2025 | — | Source ↗ | |
| 137 | Qwen3-30B-A3B-Instruct (Jul 2025) Open source Alibaba | 67.2% | — | 29 Jul 2025 | — | Source ↗ | |
| 138 | Grok-2 (Dec 2024) xAI | 65.2% | — | 12 Dec 2024 | — | Source ↗ | |
| 139 | Mistral Large 3 Open source Mistral AI | 65.1% | — | 2 Dec 2025 | — | Source ↗ | |
| 140 | DeepSeek-V3 (Mar 2025) Open source DeepSeek | 64.8% | — | 24 Mar 2025 | — | Source ↗ | |
| 141 | GPT-4o (May 2024) OpenAI | 64.5% | — | 13 May 2024 | — | Source ↗ | |
| 142 | Qwen3 14B Open source Qwen | 64.0% | — | 29 Apr 2025 | — | Source ↗ | |
| 143 | Gemini 2.0 Flash (Feb 2025) Google DeepMind,Google | 63.2% | — | 5 Feb 2025 | — | Source ↗ | |
| 144 | Qwen2.5-72B Open source Alibaba | 62.9% | — | 19 Sep 2024 | — | Source ↗ | |
| 145 | gemini-2.5-flash-lite-preview-06-17 (Default thinking length) Google DeepMind | 62.8% | — | 17 Jun 2025 | — | Source ↗ | |
| 146 | DeepSeek V3.2 Open source DeepSeek | 62.7% | — | 1 Dec 2025 | — | Source ↗ | |
| 147 | GPT-5 Nano OpenAI | 62.7% | high | 7 Aug 2025 | — | Source ↗ | |
| 148 | GPT-4 (Jun 2023) OpenAI | 62.7% | — | 13 Jun 2023 | — | Source ↗ | |
| 149 | Mistral Medium 3 Mistral | 62.3% | — | 7 May 2025 | — | Source ↗ | |
| 150 | Llama 4 Maverick Open source Meta | 61.9% | — | 6 Apr 2025 | — | Source ↗ | |
| 151 | Claude 3 Opus Anthropic | 61.6% | — | 29 Feb 2024 | — | Source ↗ | |
| 152 | GPT-4 Turbo (Nov 2023) OpenAI | 61.6% | — | 25 Jan 2024 | — | Source ↗ | |
| 153 | Llama 3.1-405B Open source Meta AI | 61.4% | — | 23 Jul 2024 | — | Source ↗ | |
| 154 | Cohere Command A Open source Cohere | 61.3% | — | 13 Mar 2025 | — | Source ↗ | |
| 155 | Magistral Small 1.2 Open source Mistral AI | 61.3% | — | 18 Sep 2025 | — | Source ↗ | |
| 156 | Mistral Large 2 (Jul 2024) Open source Mistral AI | 61.2% | — | 24 Jul 2024 | — | Source ↗ | |
| 157 | Mistral Large 2 (Nov 2024) Open source Mistral AI | 60.8% | — | 18 Nov 2024 | — | Source ↗ | |
| 158 | Qwen3 30B A3B Open source Qwen | 60.3% | — | 29 Apr 2025 | — | Source ↗ | |
| 159 | Llama 3.1-70B Open source Meta AI | 60.0% | — | 23 Jul 2024 | — | Source ↗ | |
| 160 | Mistral Small 3.2 Open source Mistral AI | 59.9% | — | 20 Jun 2025 | — | Source ↗ | |
| 161 | Qwen3 8B Open source Qwen | 59.7% | — | 28 Apr 2025 | — | Source ↗ | |
| 162 | Llama 3.3 70B Open source Meta AI | 59.5% | — | 6 Dec 2024 | — | Source ↗ | |
| 163 | Gemini 1.5 Pro (Sept 2024) Google DeepMind | 59.0% | — | 24 Sep 2024 | — | Source ↗ | |
| 164 | Mistral Small 3.1 Open source Mistral AI | 58.6% | — | 17 Mar 2025 | — | Source ↗ | |
| 165 | Llama 4 Scout Open source Meta | 57.9% | — | 5 Apr 2025 | — | Source ↗ | |
| 166 | Claude 3.5 Haiku (Oct 2024) Anthropic | 56.7% | — | 22 Oct 2024 | — | Source ↗ | |
| 167 | Gemini 1.5 Pro (May 2024) Google DeepMind | 56.5% | — | 14 May 2024 | — | Source ↗ | |
| 168 | Mistral Large mistralai | 55.9% | — | 26 Feb 2024 | — | Source ↗ | |
| 169 | Mixtral 8x22B Open source Mistral AI | 55.1% | — | 17 Apr 2024 | — | Source ↗ | |
| 170 | Command R+ Open source Cohere,Cohere Labs (formerly Cohere for AI) | 54.9% | — | 30 Aug 2024 | — | Source ↗ | |
| 171 | GPT-4o-mini OpenAI | 54.4% | — | 18 Jul 2024 | — | Source ↗ | |
| 172 | Llama 3-70B Open source Meta AI | 54.2% | — | 18 Apr 2024 | — | Source ↗ | |
| 173 | Gemini 1.5 Flash (Sep 2024) Google DeepMind | 53.8% | — | 24 Sep 2024 | — | Source ↗ | |
| 174 | Claude 3 Sonnet Anthropic | 53.6% | — | 29 Feb 2024 | — | Source ↗ | |
| 175 | Mistral Medium Mistral AI | 53.3% | — | 11 Dec 2023 | — | Source ↗ | |
| 176 | mistral-small-2402 | 53.0% | — | 26 Feb 2024 | — | Source ↗ | |
| 177 | GPT-4.1 Nano OpenAI | 52.5% | — | 14 Apr 2025 | — | Source ↗ | |
| 178 | Gemma 3 27B Open source Google | 52.5% | — | 12 Mar 2025 | — | Source ↗ | |
| 179 | Gemini 2.0 Flash-Lite Google DeepMind | 52.5% | — | 5 Feb 2025 | — | Source ↗ | |
| 180 | Claude 2 Anthropic | 51.9% | — | 11 Jul 2023 | — | Source ↗ | |
| 181 | Ministral 3B Mistral AI | 51.7% | — | 16 Oct 2024 | — | Source ↗ | |
| 182 | Gemini 1.5 Flash (May 2024) Google DeepMind | 51.4% | — | 23 May 2024 | — | Source ↗ | |
| 183 | Claude 2.1 Anthropic | 51.0% | — | 21 Nov 2023 | — | Source ↗ | |
| 184 | Gemma 3 4B Open source Google | 50.9% | — | 12 Mar 2025 | — | Source ↗ | |
| 185 | Llama 3.1-8B Open source Meta AI | 50.9% | — | 23 Jul 2024 | — | Source ↗ | |
| 186 | Claude 3 Haiku Anthropic | 50.1% | — | 7 Mar 2024 | — | Source ↗ | |
| 187 | Gemini 1.5 Flash 8B Google DeepMind | 50.0% | — | 3 Oct 2024 | — | Source ↗ | |
| 188 | Mixtral 8x7B Open source Mistral AI | 49.6% | — | 11 Dec 2023 | — | Source ↗ | |
| 189 | Mistral Small v24.09 Open source Mistral AI | 49.5% | — | 17 Sep 2024 | — | Source ↗ | |
| 190 | Gemma 3 12B Open source Google | 48.8% | — | 12 Mar 2025 | — | Source ↗ | |
| 191 | Mistral Nemo Open source Mistral | 48.6% | — | 18 Jul 2024 | — | Source ↗ | |
| 192 | GPT-3.5 Turbo (Jan 2024) OpenAI | 48.5% | — | 25 Jan 2024 | — | Source ↗ | |
| 193 | Gemma 2 27B Open source Google | 48.0% | — | 24 Jun 2024 | — | Source ↗ | |
| 194 | Qwen2.5-7B Open source Alibaba | 47.7% | — | 19 Sep 2024 | — | Source ↗ | |
| 195 | c4ai-command-r-08-2024 | 46.4% | — | 30 Aug 2024 | — | Source ↗ | |
| 196 | Gemini 1.0 Pro Google DeepMind | 45.9% | — | 13 Dec 2023 | — | Source ↗ | |
| 197 | Claude Instant Anthropic | 45.8% | — | 9 Aug 2023 | — | Source ↗ | |
| 198 | Ministral 8B Open source Mistral AI | 45.7% | — | 16 Oct 2024 | — | Source ↗ | |
| 199 | Llama 3-8B Open source Meta AI | 43.9% | — | 18 Apr 2024 | — | Source ↗ | |
| 200 | Mistral 7B v0.3 Open source Mistral AI | 42.5% | — | 27 May 2024 | — | Source ↗ | |
| 201 | Llama 2-13B Open source Meta AI | 42.2% | — | 18 Jul 2023 | — | Source ↗ | |
| 202 | Llama 2-70B Open source Meta AI | 41.6% | — | 18 Jul 2023 | — | Source ↗ |
Caveats: settings (reasoning effort, agent scaffold) differ between rows and materially affect scores; the best reported setting per model is shown. Source: Epoch AI, CC BY 4.0.