| # | Model | Score | Relative | Setting | Released | Run | Evidence |
|---|---|---|---|---|---|---|---|
| 1 | Kimi K3 (unknown) Open source Moonshot | 95.0% | unknown | 16 Jul 2026 | — | Source ↗ | |
| 2 | Claude Fable 5 Anthropic | 95.0% | high | 9 Jun 2026 | — | Source ↗ | |
| 3 | GPT-5.6 Sol (xhigh) OpenAI | 93.1% | xhigh | 9 Jul 2026 | — | Source ↗ | |
| 4 | Kimi K3 Open source MoonshotAI | 93.0% | max | 16 Jul 2026 | — | Source ↗ | |
| 5 | GPT-5.5 OpenAI | 88.1% | high | 23 Apr 2026 | — | Source ↗ | |
| 6 | Claude Opus 4.8 Anthropic | 87.5% | high | 28 May 2026 | — | Source ↗ | |
| 7 | GPT-5.6 Terra (xhigh) OpenAI | 83.8% | xhigh | 9 Jul 2026 | — | Source ↗ | |
| 8 | GPT-5.5 (medium) OpenAI | 81.3% | medium | 23 Apr 2026 | — | Source ↗ | |
| 9 | Gemini 3.8 Flash Google | 76.9% | high | 2 Sep 2026 | — | Source ↗ | |
| 10 | Grok 4.5 SpaceXAI | 74.4% | high | 8 Jul 2026 | — | Source ↗ | |
| 11 | GPT-5.6 Luna (medium) OpenAI | 61.9% | medium | 9 Jul 2026 | — | Source ↗ | |
| 12 | Claude Sonnet 5 (medium) Anthropic | 60.0% | medium | 30 Jun 2026 | — | Source ↗ | |
| 13 | Gemini 3.5 Flash Google | 58.1% | medium | 19 May 2026 | — | Source ↗ | |
| 14 | GLM 5.2 Open source Z.ai | 55.6% | high | 16 Jun 2026 | — | Source ↗ | |
| 15 | MiniMax M3 Open source MiniMax | 55.0% | — | 1 Jun 2026 | — | Source ↗ | |
| 16 | GLM 5.3 Flash Open source Z.ai | 52.5% | max | 20 Aug 2026 | — | Source ↗ | |
| 17 | Muse Spark 1.1 Meta | 52.5% | high | 9 Jul 2026 | — | Source ↗ | |
| 18 | Kimi K2.7 Code Open source MoonshotAI | 48.8% | — | 12 Jun 2026 | — | Source ↗ | |
| 19 | DeepSeek V4.1 Flash Open source DeepSeek | 46.3% | high | 9 Sep 2026 | — | Source ↗ | |
| 20 | Qwen3.8 27B (xhigh) Open source Alibaba | 45.0% | xhigh | 14 Aug 2026 | — | Source ↗ | |
| 21 | Qwen3.6 35B A3B Open source Qwen | 44.4% | — | 27 Apr 2026 | — | Source ↗ | |
| 22 | DeepSeek v4 Pro (high) Open source DeepSeek | 40.0% | high | 24 Apr 2026 | — | Source ↗ | |
| 23 | Gemma 4 31B Open source Google | 30.6% | — | 2 Apr 2026 | — | Source ↗ | |
| 24 | Mistral Medium 3.5 Open source Mistral | 26.9% | — | 28 Apr 2026 | — | Source ↗ | |
| 25 | gpt-oss-120b Open source OpenAI | 25.0% | — | 5 Aug 2025 | — | Source ↗ | |
| 26 | Laguna M.1 Poolside | 15.6% | — | 28 Apr 2026 | — | Source ↗ | |
| 27 | Trinity Large Thinking Open source Arcee AI | 15.6% | — | 1 Apr 2026 | — | 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.