Interval Quality on Sokoban
46.4Hit RateClaude Opus 4.7
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Claude Opus 4.72026.05 | 46.4 | 0.112 | |
| Claude Sonnet 4.62026.05 | 45.1 | 0.148 | |
| GPT-5.2 Instant2026.05 | 36 | 0.167 | |
| Qwen3 235B2026.05 | 10.8 | 0.029 | |
| Gemini 3.1 Pro Preview2026.05 | 8.8 | 0.313 |