| # | Model | Score | Relative | Setting | Released | Run | Evidence |
|---|---|---|---|---|---|---|---|
| 1 | GPT-6 Astra (unknown thinking) OpenAI | 20.8% | unknown | 3 Sep 2026 | — | Source ↗ | |
| 2 | Claude Fable 5.1 (unknown) Anthropic | 17.9% | unknown | 1 Sep 2026 | — | Source ↗ | |
| 3 | Claude Fable 5 Anthropic | 16.1% | — | 9 Jun 2026 | — | Source ↗ | |
| 4 | Claude Fable 5 (unknown) Anthropic | 15.8% | unknown | 9 Jun 2026 | — | Source ↗ | |
| 5 | Claude Opus 4.8 Anthropic | 8.3% | unknown | 28 May 2026 | — | Source ↗ | |
| 6 | GPT-5.5 (unknown thinking) OpenAI | 6.3% | unknown | 23 Apr 2026 | — | Source ↗ | |
| 7 | Gemini 3.7 Flash (unknown) Google DeepMind | 5.0% | unknown | 13 Aug 2026 | — | Source ↗ | |
| 8 | Claude Opus 4.6 (unknown thinking) Anthropic | 4.2% | unknown | 5 Feb 2026 | — | Source ↗ | |
| 9 | Claude Opus 4.5 (unknown thinking) Anthropic | 3.8% | unknown | 24 Nov 2025 | — | Source ↗ | |
| 10 | GPT-5.2 (medium) OpenAI | 2.5% | medium | 11 Dec 2025 | — | Source ↗ | |
| 11 | GPT-5.2 (unknown thinking) OpenAI | 2.1% | unknown | 11 Dec 2025 | — | Source ↗ | |
| 12 | Claude Sonnet 4.5 (unknown thinking) Anthropic | 2.1% | unknown | 29 Sep 2025 | — | Source ↗ | |
| 13 | GPT-5 (unknown thinking) OpenAI | 1.7% | unknown | 7 Aug 2025 | — | Source ↗ | |
| 14 | Gemini 3 Pro Preview Google DeepMind | 1.3% | — | 18 Nov 2025 | — | Source ↗ | |
| 15 | Gemini 2.5 Pro Preview (Jun 2025) Google DeepMind | 0.8% | — | 5 Jun 2025 | — | 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.