Public benchmark
Model leaderboard
Ratings from accepted blind comparisons of playable generated games.
| Rank | Model | Elo | Matches | Win rate | Status |
|---|---|---|---|---|---|
| 10 | Qwen: Qwen3.5 Plus 2026-04-20 | 1,200 | 0 | Not available | Provisional |
| 10 | Qwen: Qwen3.5-122B-A10B | 1,200 | 0 | Not available | Provisional |
| 10 | Qwen: Qwen3.5-27B | 1,200 | 0 | Not available | Provisional |
| 10 | Qwen: Qwen3.5-35B-A3B | 1,200 | 0 | Not available | Provisional |
| 10 | Qwen: Qwen3.5-9B | 1,200 | 0 | Not available | Provisional |
| 10 | Qwen: Qwen3.5-Flash | 1,200 | 0 | Not available | Provisional |
| 10 | Qwen: Qwen3.6 27B | 1,200 | 0 | Not available | Provisional |
| 10 | Qwen: Qwen3.6 35B A3B | 1,200 | 0 | Not available | Provisional |
| 10 | Qwen: Qwen3.6 Flash | 1,200 | 0 | Not available | Provisional |
| 10 | Qwen: Qwen3.6 Plus | 1,200 | 0 | Not available | Provisional |
| 10 | Qwen: Qwen3.8 2.4T A95B | 1,200 | 0 | Not available | Provisional |
| 10 | Qwen: Qwen3.8 27B (free) | 1,200 | 0 | Not available | Provisional |
| 10 | Qwen: Qwen3.8 Max Prime | 1,200 | 0 | Not available | Provisional |
| 10 | Qwen: Qwen3.8 Omni Flash | 1,200 | 0 | Not available | Provisional |
| 10 | ReMM SLERP 13B | 1,200 | 0 | Not available | Provisional |
| 10 | Reflection: Beam 1.0 501B A23B | 1,200 | 0 | Not available | Provisional |
| 10 | Reka Edge | 1,200 | 0 | Not available | Provisional |
| 10 | Reka Flash 3 | 1,200 | 0 | Not available | Provisional |
| 10 | Relace: Relace Apply 3 | 1,200 | 0 | Not available | Provisional |
| 10 | Relace: Relace Search | 1,200 | 0 | Not available | Provisional |
| 10 | Sakana: Fugu Max | 1,200 | 0 | Not available | Provisional |
| 10 | Sakana: Fugu Ultra v2 | 1,200 | 0 | Not available | Provisional |
| 10 | Sakana: Sakana Namazu | 1,200 | 0 | Not available | Provisional |
| 10 | Sao10K: Llama 3 8B Lunaris | 1,200 | 0 | Not available | Provisional |
| 10 | Sao10K: Llama 3.1 Euryale 70B v2.2 | 1,200 | 0 | Not available | Provisional |