Updated daily at 05:00 UTC
Benchmark
Mystery Game Puzzles
Measures: 100 mid-game positions from a puzzle variant of a well-known game whose identity is kept secret; the model must name the single best next move.
- Maintainer
- Epoch AI
- Introduced
- 2026
- Status
- Active
- Saturation
- Near-ceiling
- Typical provenance
- Independent
- In scoring set
- No
Leaderboard
Independent and third-party results only.
| Model | Puzzles solved | Provenance | Source | Evaluated |
|---|---|---|---|---|
| GPT-6 Astra OpenAI · Max effort | 84.0% | Independent | Epoch AI | 30 Aug 2026 |
| Claude Opus 5 Anthropic · Max effort | 59.0% | Independent | Epoch AI | 25 Jul 2026 |
| Claude Fable 5.1 Anthropic · Max effort | 58.0% | Independent | Epoch AI | 1 Sep 2026 |
| GPT-5.6 Sol OpenAI · Max effort | 58.0% | Independent | Epoch AI | 28 Jul 2026 |
| GPT-5.5 OpenAI · Xhigh effort | 56.0% | Independent | Epoch AI | 24 Jul 2026 |
| Claude Fable 5 Anthropic · Max effort | 52.0% | Independent | Epoch AI | 31 Jul 2026 |
| GPT-5.5 OpenAI · High effort | 52.0% | Independent | Epoch AI | 27 Jul 2026 |
| Gemini 3.8 Flash Google DeepMind · High effort | 47.0% | Independent | Epoch AI | 2 Sep 2026 |
| DeepSeek-V4-Pro DeepSeek · Max effort | 43.0% | Independent | Epoch AI | 19 Aug 2026 |
| Qwen3.8-Max Alibaba Group · Xhigh effort | 38.0% | Independent | Epoch AI | 5 Aug 2026 |
| Claude Opus 5 | 37.0% | Independent | Epoch AI | 6 Aug 2026 |
| Gemini 3.7 Flash Google DeepMind · High effort | 37.0% | Independent | Epoch AI | 14 Aug 2026 |
| GPT-5.4 OpenAI · Xhigh effort | 37.0% | Independent | Epoch AI | 24 Jul 2026 |
| Claude Opus 4.8 Anthropic · Max effort | 36.0% | Independent | Epoch AI | 25 Jul 2026 |
| Claude Sonnet 5 Anthropic · Max effort | 35.0% | Independent | Epoch AI | 28 Jul 2026 |
| GPT-5.6 Terra OpenAI · Max effort | 35.0% | Independent | Epoch AI | 28 Jul 2026 |
| DeepSeek-V4-Flash DeepSeek · Max effort | 34.0% | Independent | Epoch AI | 5 Aug 2026 |
| Gemini 3.1 Pro Google DeepMind · High effort | 34.0% | Independent | Epoch AI | 27 Jul 2026 |
| Grok 4.6 SpaceXAI (xAI) · Xhigh effort | 34.0% | Independent | Epoch AI | 14 Aug 2026 |
| GLM-5.3 Z.ai (Zhipu AI) · Max effort | 33.0% | Independent | Epoch AI | 30 Aug 2026 |
| GPT-5.6 Sol OpenAI · None effort | 33.0% | Independent | Epoch AI | 27 Aug 2026 |
| Gemini 3.1 Pro Google DeepMind · Medium effort | 32.0% | Independent | Epoch AI | 6 Aug 2026 |
| Gemini 3.5 Flash Google DeepMind · High effort | 32.0% | Independent | Epoch AI | 27 Jul 2026 |
| Qwen3.7-Max | 32.0% | Independent | Epoch AI | 28 Jul 2026 |
| Claude Opus 4.8 Anthropic · Xhigh effort | 31.0% | Independent | Epoch AI | 26 Jul 2026 |
| Gemini 3.6 Flash Google DeepMind · High effort | 30.0% | Independent | Epoch AI | 5 Aug 2026 |
| Gemini 3.1 Pro Google DeepMind · Low effort | 29.0% | Independent | Epoch AI | 6 Aug 2026 |
| o3 OpenAI · High effort | 29.0% | Independent | Epoch AI | 27 Aug 2026 |
| Claude Opus 4.7 Anthropic · Max effort | 28.0% | Independent | Epoch AI | 25 Jul 2026 |
| Gemini 3.5 Flash Google DeepMind · Low effort | 28.0% | Independent | Epoch AI | 5 Aug 2026 |
| GPT-5.4 OpenAI · Medium effort | 28.0% | Independent | Epoch AI | 28 Aug 2026 |
| GPT-5.5 OpenAI · Low effort | 28.0% | Independent | Epoch AI | 28 Aug 2026 |
| Gemini 3 Flash Google DeepMind · Low effort | 26.0% | Independent | Epoch AI | 5 Aug 2026 |
| GPT-5.6 Sol OpenAI · Low effort | 26.0% | Independent | Epoch AI | 27 Aug 2026 |
| Kimi K3 Moonshot AI · Max effort | 26.0% | Independent | Epoch AI | 29 Jul 2026 |
| Qwen3.7-Max Alibaba Group · None effort | 26.0% | Independent | Epoch AI | 27 Aug 2026 |
| Claude Opus 4.6 Anthropic · Max effort | 25.0% | Independent | Epoch AI | 25 Jul 2026 |
| Gemini 3 Flash Google DeepMind · Minimal effort | 25.0% | Independent | Epoch AI | 5 Aug 2026 |
| Gemini 3.6 Flash Google DeepMind · Minimal effort | 25.0% | Independent | Epoch AI | 5 Aug 2026 |
| Muse Spark 1.3 Meta · Max effort | 25.0% | Independent | Epoch AI | 18 Sep 2026 |
| Gemini 3.6 Flash Google DeepMind · Low effort | 23.0% | Independent | Epoch AI | 5 Aug 2026 |
| GPT-5 OpenAI · High effort | 23.0% | Independent | Epoch AI | 5 Aug 2026 |
| GPT-5.2 OpenAI · High effort | 23.0% | Independent | Epoch AI | 6 Aug 2026 |
| o3 OpenAI · Medium effort | 23.0% | Independent | Epoch AI | 27 Aug 2026 |
| Claude Opus 4.5 Anthropic · 48K effort | 22.0% | Independent | Epoch AI | 25 Jul 2026 |
| GPT-5.2 OpenAI · Medium effort | 22.0% | Independent | Epoch AI | 28 Aug 2026 |
| Qwen3.6-35B-A3B Alibaba Group · None effort | 22.0% | Independent | Epoch AI | 27 Aug 2026 |
| Claude Opus 4.1 Anthropic · 24K effort | 21.0% | Independent | Epoch AI | 25 Jul 2026 |
| GPT-5.6 Luna OpenAI · Max effort | 21.0% | Independent | Epoch AI | 28 Jul 2026 |
| Gemini 3 Flash Google DeepMind · High effort | 20.0% | Independent | Epoch AI | 5 Aug 2026 |
| GPT-5.6 Luna OpenAI · None effort | 20.0% | Independent | Epoch AI | 30 Aug 2026 |
| Qwen3.5-Flash | 20.0% | Independent | Epoch AI | 27 Aug 2026 |
| Gemini 3.5 Flash-Lite Google DeepMind · Low effort | 19.0% | Independent | Epoch AI | 5 Aug 2026 |
| GLM-5.2 Z.ai (Zhipu AI) · Low effort | 19.0% | Independent | Epoch AI | 27 Aug 2026 |
| GPT-5.1 OpenAI · Low effort | 19.0% | Independent | Epoch AI | 27 Aug 2026 |
| GPT-5.6 Terra OpenAI · Low effort | 19.0% | Independent | Epoch AI | 27 Aug 2026 |
| Qwen3.6-Max-Preview Alibaba Group · None effort | 19.0% | Independent | Epoch AI | 27 Aug 2026 |
| GLM-5.2 Z.ai (Zhipu AI) · Minimal effort | 18.0% | Independent | Epoch AI | 27 Aug 2026 |
| GLM-5.2 Z.ai (Zhipu AI) · None effort | 18.0% | Independent | Epoch AI | 27 Aug 2026 |
| GPT-5.5 OpenAI · None effort | 18.0% | Independent | Epoch AI | 27 Aug 2026 |
| Kimi K2.6 | 18.0% | Independent | Epoch AI | 17 Jul 2026 |
| Qwen3.5-397B-A17B Alibaba Group · None effort | 18.0% | Independent | Epoch AI | 27 Aug 2026 |
| Qwen3.6-Flash Alibaba Group · None effort | 18.0% | Independent | Epoch AI | 27 Aug 2026 |
| Claude Sonnet 4.5 Anthropic · 48K effort | 17.0% | Independent | Epoch AI | 25 Jul 2026 |
| DeepSeek-V4-Pro DeepSeek · None effort | 17.0% | Independent | Epoch AI | 27 Aug 2026 |
| GPT-5.4 OpenAI · Low effort | 17.0% | Independent | Epoch AI | 27 Aug 2026 |
| GPT-5.6 Luna OpenAI · Low effort | 17.0% | Independent | Epoch AI | 27 Aug 2026 |
| Qwen3.5-122B-A10B Alibaba Group · None effort | 17.0% | Independent | Epoch AI | 27 Aug 2026 |
| Qwen3.5-Plus Alibaba Group · None effort | 17.0% | Independent | Epoch AI | 27 Aug 2026 |
| Qwen3.7-Plus | 17.0% | Independent | Epoch AI | 28 Aug 2026 |
| Qwen3.7-Plus Alibaba Group · None effort | 17.0% | Independent | Epoch AI | 27 Aug 2026 |
| Claude Sonnet 4.6 Anthropic · Low effort | 16.0% | Independent | Epoch AI | 6 Aug 2026 |
| Claude Sonnet 5 | 16.0% | Independent | Epoch AI | 6 Aug 2026 |
| GPT-5 OpenAI · Low effort | 16.0% | Independent | Epoch AI | 27 Aug 2026 |
| GPT-5.1 OpenAI · Medium effort | 16.0% | Independent | Epoch AI | 27 Aug 2026 |
| GPT-5.4 OpenAI · None effort | 16.0% | Independent | Epoch AI | 30 Aug 2026 |
| Qwen3.5-Flash Alibaba Group · None effort | 16.0% | Independent | Epoch AI | 27 Aug 2026 |
| Qwen3.5-Plus | 16.0% | Independent | Epoch AI | 5 Aug 2026 |
| Claude Opus 4.6 | 15.0% | Independent | Epoch AI | 6 Aug 2026 |
| GLM-5.2 Z.ai (Zhipu AI) · Medium effort | 15.0% | Independent | Epoch AI | 27 Aug 2026 |
| GPT-5 OpenAI · Medium effort | 15.0% | Independent | Epoch AI | 27 Aug 2026 |
| GPT-5.1 OpenAI · None effort | 15.0% | Independent | Epoch AI | 27 Aug 2026 |
| Qwen3.7-Flash | 15.0% | Independent | Epoch AI | 28 Aug 2026 |
| Muse Spark 1.3 Meta · Xhigh effort | 14.1% | Independent | Epoch AI | 16 Sep 2026 |
| Claude Sonnet 4.6 | 14.0% | Independent | Epoch AI | 6 Aug 2026 |
| GPT-5 OpenAI · Minimal effort | 14.0% | Independent | Epoch AI | 27 Aug 2026 |
| GPT-5.2 OpenAI · None effort | 14.0% | Independent | Epoch AI | 27 Aug 2026 |
| GPT-5.6 Terra OpenAI · None effort | 14.0% | Independent | Epoch AI | 27 Aug 2026 |
| Qwen3.7-Flash Alibaba Group · None effort | 14.0% | Independent | Epoch AI | 27 Aug 2026 |
| Claude Opus 4.7 | 13.0% | Independent | Epoch AI | 6 Aug 2026 |
| Gemini 3.5 Flash-Lite Google DeepMind · Minimal effort | 12.0% | Independent | Epoch AI | 5 Aug 2026 |
| GPT-4 | 12.0% | Independent | Epoch AI | 28 Aug 2026 |
| GPT-4o mini | 12.0% | Independent | Epoch AI | 27 Aug 2026 |
| GPT-5.6 Luna OpenAI · Medium effort | 12.0% | Independent | Epoch AI | 27 Aug 2026 |
| Kimi K2.6 Moonshot AI · None effort | 12.0% | Independent | Epoch AI | 29 Aug 2026 |
| Kimi K3 Moonshot AI · Low effort | 12.0% | Independent | Epoch AI | 29 Aug 2026 |
| Qwen3.6-Flash | 12.0% | Independent | Epoch AI | 27 Aug 2026 |
| Qwen3.6-Plus Alibaba Group · None effort | 12.0% | Independent | Epoch AI | 27 Aug 2026 |
| GPT-5.4 mini OpenAI · None effort | 11.0% | Independent | Epoch AI | 27 Aug 2026 |
| GPT-5 mini OpenAI · Minimal effort | 10.0% | Independent | Epoch AI | 27 Aug 2026 |
| GPT-5.2 OpenAI · Low effort | 10.0% | Independent | Epoch AI | 27 Aug 2026 |
| GPT-5.6 Terra OpenAI · Medium effort | 10.0% | Independent | Epoch AI | 28 Aug 2026 |
| GPT-5 nano OpenAI · Medium effort | 9.0% | Independent | Epoch AI | 27 Aug 2026 |
| GPT-5 nano OpenAI · Low effort | 9.0% | Independent | Epoch AI | 27 Aug 2026 |
| GPT-5.4 nano OpenAI · None effort | 9.0% | Independent | Epoch AI | 27 Aug 2026 |
| Qwen3-235B-A22B-Thinking-2507 | 9.0% | Independent | Epoch AI | 27 Aug 2026 |
| GLM-5.3-Flash Z.ai (Zhipu AI) · Max effort | 8.0% | Independent | Epoch AI | 28 Aug 2026 |
| GPT-5 nano OpenAI · High effort | 8.0% | Independent | Epoch AI | 27 Aug 2026 |
| GPT-5.4 mini OpenAI · Low effort | 8.0% | Independent | Epoch AI | 27 Aug 2026 |
| MiniMax-M3 MiniMax · None effort | 8.0% | Independent | Epoch AI | 27 Aug 2026 |
| Claude Opus 4.6 Anthropic · Low effort | 7.0% | Independent | Epoch AI | 6 Aug 2026 |
| GPT-4.1 mini | 7.0% | Independent | Epoch AI | 27 Aug 2026 |
| GPT-5.4 mini OpenAI · Medium effort | 7.0% | Independent | Epoch AI | 27 Aug 2026 |
| MiniMax-M3 | 7.0% | Independent | Epoch AI | 27 Aug 2026 |
| o3-mini OpenAI · High effort | 7.0% | Independent | Epoch AI | 27 Aug 2026 |
| Qwen3.6-27B Alibaba Group · None effort | 7.0% | Independent | Epoch AI | 27 Aug 2026 |
| GPT-5.4 nano OpenAI · Medium effort | 6.0% | Independent | Epoch AI | 27 Aug 2026 |
| Inkling Small Thinking Machines Lab · Xhigh effort | 6.0% | Independent | Epoch AI | 15 Aug 2026 |
| GPT-5 mini OpenAI · High effort | 5.0% | Independent | Epoch AI | 27 Aug 2026 |
| GPT-5 nano OpenAI · Minimal effort | 5.0% | Independent | Epoch AI | 27 Aug 2026 |
| GPT-5.4 nano OpenAI · High effort | 5.0% | Independent | Epoch AI | 27 Aug 2026 |
| o4-mini OpenAI · High effort | 5.0% | Independent | Epoch AI | 27 Aug 2026 |
| Qwen3-Max | 5.0% | Independent | Epoch AI | 27 Aug 2026 |
| GPT-5 mini OpenAI · Medium effort | 4.0% | Independent | Epoch AI | 27 Aug 2026 |
| GPT-3.5 Turbo | 3.0% | Independent | Epoch AI | 27 Aug 2026 |
| GPT-5.4 nano OpenAI · Low effort | 3.0% | Independent | Epoch AI | 27 Aug 2026 |
| gpt-oss-120b OpenAI · Medium effort | 2.0% | Independent | Epoch AI | 27 Aug 2026 |
| gpt-oss-120b OpenAI · High effort | 0.0% | Independent | Epoch AI | 27 Aug 2026 |
- Epoch AI, 'Capabilities & benchmarking', epoch.ai/benchmarks, CC BY 4.0.