Public benchmark
Model leaderboard
Ratings from accepted blind comparisons of playable generated games.
| Rank | Model | Elo | Matches | Win rate | Status |
|---|---|---|---|---|---|
| 10 | qwen/qwen3.7-flash | 1,200 | 0 | Not available | Provisional |
| 10 | qwen/qwen3.7-max | 1,200 | 0 | Not available | Provisional |
| 10 | qwen/qwen3.7-plus | 1,200 | 0 | Not available | Provisional |
| 10 | qwen/qwen3.8-27b | 1,200 | 0 | Not available | Provisional |
| 10 | qwen/qwen3.8-flash | 1,200 | 0 | Not available | Provisional |
| 10 | sakana/fugu-ultra | 1,200 | 0 | Not available | Provisional |
| 10 | stealth/ox-alpha | 1,200 | 0 | Not available | Provisional |
| 10 | tencent/hy3 | 1,200 | 0 | Not available | Provisional |
| 10 | tencent/hy3-preview:free | 1,200 | 0 | Not available | Provisional |
| 10 | thinkingmachines/inkling | 1,200 | 0 | Not available | Provisional |
| 10 | x-ai/grok-4-07-09 | 1,200 | 0 | Not available | Provisional |
| 10 | x-ai/grok-4-fast:free | 1,200 | 0 | Not available | Provisional |
| 10 | x-ai/grok-4.20-multi-agent | 1,200 | 0 | Not available | Provisional |
| 10 | x-ai/grok-4.5 | 1,200 | 0 | Not available | Provisional |
| 10 | x-ai/grok-4.6 | 1,200 | 0 | Not available | Provisional |
| 10 | x-ai/grok-code-fast-1 | 1,200 | 0 | Not available | Provisional |
| 10 | xAI: Grok 3 Beta | 1,200 | 0 | Not available | Provisional |
| 10 | z-ai/glm-4.7 | 1,200 | 0 | Not available | Provisional |
| 10 | z-ai/glm-5 | 1,200 | 0 | Not available | Provisional |
| 10 | z-ai/glm-5.1 | 1,200 | 0 | Not available | Provisional |
| 10 | z-ai/glm-5.2 | 1,200 | 0 | Not available | Provisional |
| 10 | z-ai/glm-5.3 | 1,200 | 0 | Not available | Provisional |
| 10 | z-ai/glm-5.3-flash | 1,200 | 0 | Not available | Provisional |
| 10 | zenith | 1,200 | 0 | Not available | Provisional |
| 100 | OpenAI: o3 | 1,184 | 1 | 0.0% | Provisional |