Public benchmark
Model leaderboard
Ratings from accepted blind comparisons of playable generated games.
| Rank | Model | Elo | Matches | Win rate | Status |
|---|---|---|---|---|---|
| 10 | OpenAI: GPT-5 Nano (batch) | 1,200 | 0 | Not available | Provisional |
| 10 | OpenAI: GPT-5 Pro (batch) | 1,200 | 0 | Not available | Provisional |
| 10 | OpenAI: GPT-5.1 | 1,200 | 0 | Not available | Provisional |
| 10 | OpenAI: GPT-5.1 (batch) | 1,200 | 0 | Not available | Provisional |
| 10 | OpenAI: GPT-5.1-Codex-Max | 1,200 | 0 | Not available | Provisional |
| 10 | OpenAI: GPT-5.1-Codex-Mini | 1,200 | 0 | Not available | Provisional |
| 10 | OpenAI: GPT-5.2 | 1,200 | 0 | Not available | Provisional |
| 10 | OpenAI: GPT-5.2 (batch) | 1,200 | 0 | Not available | Provisional |
| 10 | OpenAI: GPT-5.2 Chat | 1,200 | 0 | Not available | Provisional |
| 10 | OpenAI: GPT-5.2 Pro (batch) | 1,200 | 0 | Not available | Provisional |
| 10 | OpenAI: GPT-5.3-Codex | 1,200 | 0 | Not available | Provisional |
| 10 | OpenAI: GPT-5.4 (batch) | 1,200 | 0 | Not available | Provisional |
| 10 | OpenAI: GPT-5.4 Image 2 | 1,200 | 0 | Not available | Provisional |
| 10 | OpenAI: GPT-5.4 Mini | 1,200 | 0 | Not available | Provisional |
| 10 | OpenAI: GPT-5.4 Mini (batch) | 1,200 | 0 | Not available | Provisional |
| 10 | OpenAI: GPT-5.4 Nano | 1,200 | 0 | Not available | Provisional |
| 10 | OpenAI: GPT-5.4 Nano (batch) | 1,200 | 0 | Not available | Provisional |
| 10 | OpenAI: GPT-5.4 Pro | 1,200 | 0 | Not available | Provisional |
| 10 | OpenAI: GPT-5.4 Pro (batch) | 1,200 | 0 | Not available | Provisional |
| 10 | OpenAI: GPT-5.5 (batch) | 1,200 | 0 | Not available | Provisional |
| 10 | OpenAI: GPT-5.5 Pro | 1,200 | 0 | Not available | Provisional |
| 10 | OpenAI: GPT-5.5 Pro (batch) | 1,200 | 0 | Not available | Provisional |
| 10 | OpenAI: GPT-5.6 Luna (batch) | 1,200 | 0 | Not available | Provisional |
| 10 | OpenAI: GPT-5.6 Luna Pro | 1,200 | 0 | Not available | Provisional |
| 10 | OpenAI: GPT-5.6 Luna Pro (batch) | 1,200 | 0 | Not available | Provisional |