Sokoban on Sokoban (Hard-2)
87.89AccuracyGPT-5
Evaluation Results
| Method | Links | |
|---|---|---|
| GPT-5category=Proprietary Models2025.11 | 87.89 | |
| OpenAI o3category=Proprietary Models2025.11 | 85.16 | |
| OpenAI o4-minicategory=Proprietary Models2025.11 | 62.89 | |
| WMAct2025.11 | 49.9 | |
| PPO - Interactivemode=interactive2025.11 | 46.83 | |
| Claude 4.5 Sonnetcategory=Proprietary Models2025.11 | 29.3 | |
| Gemini 2.5 Procategory=Proprietary Models2025.11 | 20.31 | |
| Qwen3-14Bcategory=Opensource Models2025.11 | 5.08 | |
| Qwen3-8B-Ownbackbone=Qwen3-8B2025.11 | 1.39 | |
| Qwen3-8Bcategory=Opensource Models2025.11 | 1.17 | |
| PPO - EntirePlanmode=single-turn output2025.11 | 0.35 | |
| GPT-4ocategory=Proprietary Models2025.11 | 0 | |
| Qwen2.5-7B-Instructcategory=Opensource Models2025.11 | 0 | |
| Qwen2.5-32B-Instructcategory=Opensource Models2025.11 | 0 |