| # | Model | Score | Relative | Setting | Released | Run | Evidence |
|---|---|---|---|---|---|---|---|
| 1 | Claude Fable 5 Anthropic | 81.9% | max | 9 Jun 2026 | — | Source ↗ | |
| 2 | Claude Opus 5 (unknown) Anthropic | 80.6% | unknown | 24 Jul 2026 | — | Source ↗ | |
| 3 | Gemini 3.1 Pro Preview Google | 79.6% | — | 19 Feb 2026 | — | Source ↗ | |
| 4 | GPT-5.5 Pro OpenAI | 76.9% | high | 23 Apr 2026 | — | Source ↗ | |
| 5 | GPT-5.5 Pro (unknown thinking) OpenAI | 76.9% | unknown | 23 Apr 2026 | — | Source ↗ | |
| 6 | Gemini 3.5 Flash Google | 76.7% | unknown | 19 May 2026 | — | Source ↗ | |
| 7 | Gemini 3 Pro Preview Google DeepMind | 76.4% | — | 18 Nov 2025 | — | Source ↗ | |
| 8 | Grok 4.6 (unknown) xAI | 75.9% | unknown | 12 Aug 2026 | — | Source ↗ | |
| 9 | Muse Spark 1.2 Meta | 74.5% | — | 5 Aug 2026 | — | Source ↗ | |
| 10 | GPT-5.4 Pro OpenAI | 74.1% | unknown | 5 Mar 2026 | — | Source ↗ | |
| 11 | GPT-5.6 Sol (pro, xhigh) OpenAI | 71.7% | proxhigh | 9 Jul 2026 | — | Source ↗ | |
| 12 | GPT-5.6 Sol (pro, unknown thinking) OpenAI | 71.7% | prounknown | 9 Jul 2026 | — | Source ↗ | |
| 13 | Qwen3.7 Max Qwen | 70.4% | — | 19 May 2026 | — | Source ↗ | |
| 14 | Grok 4.5 (unknown thinking) xAI | 70.0% | unknown | 8 Jul 2026 | — | Source ↗ | |
| 15 | GPT-5.5 (unknown thinking) OpenAI | 69.0% | unknown | 23 Apr 2026 | — | Source ↗ | |
| 16 | Claude Opus 4.6 (no thinking) Anthropic | 67.6% | — | 5 Feb 2026 | — | Source ↗ | |
| 17 | GPT-5.6 Sol (xhigh) OpenAI | 64.8% | xhigh | 9 Jul 2026 | — | Source ↗ | |
| 18 | GPT-5.6 Sol (unknown thinking) OpenAI | 64.8% | unknown | 9 Jul 2026 | — | Source ↗ | |
| 19 | Claude Opus 4.8 Anthropic | 64.8% | unknown | 28 May 2026 | — | Source ↗ | |
| 20 | Qwen 3.6 Max (Preview) Alibaba | 63.0% | — | 20 Apr 2026 | — | Source ↗ | |
| 21 | Gemini 2.5 Pro Preview (Jun 2025) Google DeepMind | 62.4% | — | 5 Jun 2025 | — | Source ↗ | |
| 22 | Claude Opus 4.5 (no thinking) Anthropic | 62.0% | — | 24 Nov 2025 | — | Source ↗ | |
| 23 | Claude Opus 4.5 (unknown thinking) Anthropic | 62.0% | unknown | 24 Nov 2025 | — | Source ↗ | |
| 24 | Claude Opus 4.7 (unknown) Anthropic | 61.7% | unknown | 16 Apr 2026 | — | Source ↗ | |
| 25 | GPT-5 Pro OpenAI | 61.6% | high | 7 Oct 2025 | — | Source ↗ | |
| 26 | DeepSeek V4 Flash 0731 (unknown) Open source DeepSeek | 61.1% | unknown | 31 Jul 2026 | — | Source ↗ | |
| 27 | Gemini 3 Flash Preview Google | 61.1% | — | 17 Dec 2025 | — | Source ↗ | |
| 28 | Kimi K3 Open source MoonshotAI | 60.7% | max | 16 Jul 2026 | — | Source ↗ | |
| 29 | Claude Sonnet 5 (unknown thinking) Anthropic | 60.6% | unknown | 30 Jun 2026 | — | Source ↗ | |
| 30 | Grok 4 xAI | 60.5% | — | 9 Jul 2025 | — | Source ↗ | |
| 31 | Claude Opus 4.1 Anthropic | 60.0% | — | 5 Aug 2025 | — | Source ↗ | |
| 32 | Claude Opus 4.1 (unknown thinking) Anthropic | 60.0% | unknown | 5 Aug 2025 | — | Source ↗ | |
| 33 | GLM 5.2 Open source Z.ai | 58.8% | unknown | 16 Jun 2026 | — | Source ↗ | |
| 34 | Claude Opus 4 Anthropic | 58.8% | 12K | 22 May 2025 | — | Source ↗ | |
| 35 | Kimi K2.7 Code Open source MoonshotAI | 57.9% | — | 12 Jun 2026 | — | Source ↗ | |
| 36 | GPT-5.2 Pro OpenAI | 57.4% | — | 11 Dec 2025 | — | Source ↗ | |
| 37 | GPT-5 OpenAI | 56.7% | high | 7 Aug 2025 | — | Source ↗ | |
| 38 | Grok 4.1 Fast xAI | 56.0% | — | 19 Nov 2025 | — | Source ↗ | |
| 39 | GLM 5.1 Open source Z.ai | 55.1% | — | 7 Apr 2026 | — | Source ↗ | |
| 40 | Claude Sonnet 4.5 (unknown thinking) Anthropic | 54.3% | unknown | 29 Sep 2025 | — | Source ↗ | |
| 41 | Claude Sonnet 4.5 (12k thinking) Anthropic | 54.3% | 12K | 29 Sep 2025 | — | Source ↗ | |
| 42 | GLM 5 Open source Z.ai | 53.2% | — | 11 Feb 2026 | — | Source ↗ | |
| 43 | GPT-5.1 OpenAI | 53.2% | high | 13 Nov 2025 | — | Source ↗ | |
| 44 | o3 OpenAI | 53.1% | high | 16 Apr 2025 | — | Source ↗ | |
| 45 | DeepSeek-V3.2-Speciale Open source DeepSeek | 52.6% | — | 1 Dec 2025 | — | Source ↗ | |
| 46 | Gemini 2.5 Pro Preview (Mar 2025) Google DeepMind | 51.6% | — | 31 Mar 2025 | — | Source ↗ | |
| 47 | Gemini 2.5 Pro Exp (Mar 2025) Google DeepMind | 51.6% | — | 25 Mar 2025 | — | Source ↗ | |
| 48 | Inkling Open source Thinking Machines | 50.0% | — | 15 Jul 2026 | — | Source ↗ | |
| 49 | GPT-5.6 Terra (xhigh) OpenAI | 48.9% | xhigh | 9 Jul 2026 | — | Source ↗ | |
| 50 | GPT-5.6 Terra (unknown thinking) OpenAI | 48.9% | unknown | 9 Jul 2026 | — | Source ↗ | |
| 51 | GLM-4.7 (Novita) Open source Z.ai (Zhipu AI) | 47.7% | — | 22 Dec 2025 | — | Source ↗ | |
| 52 | GLM 4.7 Open source Z.ai | 47.7% | — | 22 Dec 2025 | — | Source ↗ | |
| 53 | GPT-5.6 Luna (unknown thinking) OpenAI | 46.8% | unknown | 9 Jul 2026 | — | Source ↗ | |
| 54 | GPT-5.6 Luna (xhigh) OpenAI | 46.8% | xhigh | 9 Jul 2026 | — | Source ↗ | |
| 55 | Kimi K2.5 Open source MoonshotAI | 46.8% | — | 27 Jan 2026 | — | Source ↗ | |
| 56 | Claude 3.7 Sonnet (12k thinking) Anthropic | 46.4% | 12K | 24 Feb 2025 | — | Source ↗ | |
| 57 | Claude 3.7 Sonnet (unknown thinking) Anthropic | 46.4% | unknown | 24 Feb 2025 | — | Source ↗ | |
| 58 | DeepSeek v4 (unknown) Open source DeepSeek | 46.3% | unknown | 24 Apr 2026 | — | Source ↗ | |
| 59 | GPT-5.2 OpenAI | 45.8% | high | 11 Dec 2025 | — | Source ↗ | |
| 60 | MiniMax M3 Open source MiniMax | 45.8% | — | 1 Jun 2026 | — | Source ↗ | |
| 61 | GPT-5.2 (unknown thinking) OpenAI | 45.8% | unknown | 11 Dec 2025 | — | Source ↗ | |
| 62 | Claude Sonnet 4 Anthropic | 45.5% | 12K | 22 May 2025 | — | Source ↗ | |
| 63 | Claude Sonnet 4 (unknown thinking) Anthropic | 45.5% | unknown | 22 May 2025 | — | Source ↗ | |
| 64 | Claude 3.7 Sonnet Anthropic | 44.9% | — | 24 Feb 2025 | — | Source ↗ | |
| 65 | o1-preview OpenAI | 41.7% | — | 12 Sep 2024 | — | Source ↗ | |
| 66 | Claude 3.5 Sonnet (Oct 2024) Anthropic | 41.4% | — | 22 Oct 2024 | — | Source ↗ | |
| 67 | Gemini 2.5 Flash (Jun 2025) Google DeepMind | 41.2% | — | 17 Jun 2025 | — | Source ↗ | |
| 68 | DeepSeek-R1 (May 2025) Open source DeepSeek | 40.8% | — | 28 May 2025 | — | Source ↗ | |
| 69 | o1 OpenAI | 40.1% | high | 17 Dec 2024 | — | Source ↗ | |
| 70 | DeepSeek V3.1 Open source DeepSeek | 40.0% | — | 21 Aug 2025 | — | Source ↗ | |
| 71 | o4 Mini OpenAI | 38.7% | high | 16 Apr 2025 | — | Source ↗ | |
| 72 | o1 (medium) OpenAI | 36.7% | medium | 17 Dec 2024 | — | Source ↗ | |
| 73 | Grok 3 xAI | 36.1% | — | 9 Apr 2025 | — | Source ↗ | |
| 74 | Qwen 3.6 Flash (2026-04-16) Alibaba | 35.2% | — | 27 Apr 2026 | — | Source ↗ | |
| 75 | GPT-4.5 Preview (Feb 2025) OpenAI | 34.5% | — | 27 Feb 2025 | — | Source ↗ | |
| 76 | Gemini 2.0 Flash (Dec 2024) Google DeepMind,Google | 31.1% | — | 11 Dec 2024 | — | Source ↗ | |
| 77 | Qwen3 235B A22B Open source Qwen | 31.0% | — | 28 Apr 2025 | — | Source ↗ | |
| 78 | R1 Open source DeepSeek | 30.9% | — | 20 Jan 2025 | — | Source ↗ | |
| 79 | Gemini 2.0 Flash Thinking Exp Google DeepMind,Google | 30.7% | — | 21 Jan 2025 | — | Source ↗ | |
| 80 | Llama 4 Maverick Open source Meta | 27.7% | — | 6 Apr 2025 | — | Source ↗ | |
| 81 | Claude 3.5 Sonnet (Jun 2024) Anthropic | 27.5% | — | 20 Jun 2024 | — | Source ↗ | |
| 82 | DeepSeek-V3 (Mar 2025) Open source DeepSeek | 27.2% | — | 24 Mar 2025 | — | Source ↗ | |
| 83 | Gemini 1.5 Pro (Sept 2024) Google DeepMind | 27.1% | — | 24 Sep 2024 | — | Source ↗ | |
| 84 | GPT-4.1 OpenAI | 27.0% | — | 14 Apr 2025 | — | Source ↗ | |
| 85 | Kimi K2 Instruct Open source Moonshot | 26.3% | — | 12 Jul 2025 | — | Source ↗ | |
| 86 | GPT-4 Turbo (Apr 2024) OpenAI | 25.1% | — | 9 Apr 2024 | — | Source ↗ | |
| 87 | Claude 3 Opus Anthropic | 23.5% | — | 29 Feb 2024 | — | Source ↗ | |
| 88 | Llama 3.1-405B Open source Meta AI | 23.0% | — | 23 Jul 2024 | — | Source ↗ | |
| 89 | o3 Mini OpenAI | 22.8% | high | 31 Jan 2025 | — | Source ↗ | |
| 90 | Grok-2 (Dec 2024) xAI | 22.7% | — | 12 Dec 2024 | — | Source ↗ | |
| 91 | Mistral Large 2 (Jul 2024) Open source Mistral AI | 22.5% | — | 24 Jul 2024 | — | Source ↗ | |
| 92 | gpt-oss-120b Open source OpenAI | 22.1% | — | 5 Aug 2025 | — | Source ↗ | |
| 93 | Llama 3.3 70B Open source Meta AI | 19.9% | — | 6 Dec 2024 | — | Source ↗ | |
| 94 | DeepSeek V3 Open source DeepSeek | 18.9% | — | 26 Dec 2024 | — | Source ↗ | |
| 95 | o1-mini (medium) OpenAI | 18.1% | medium | 12 Sep 2024 | — | Source ↗ | |
| 96 | GPT-4o (Aug 2024) OpenAI | 17.8% | — | 6 Aug 2024 | — | Source ↗ | |
| 97 | Command R+ Open source Cohere,Cohere Labs (formerly Cohere for AI) | 17.4% | — | 30 Aug 2024 | — | Source ↗ | |
| 98 | GPT-4o-mini OpenAI | 10.7% | — | 18 Jul 2024 | — | 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.