Cursorbench benchmark scores from Epoch AI's benchmark data collection.
Sourced from Epoch AI's systematic benchmark data collection covering frontier models.
Auto-generated definition — editorial description not yet written.
Higher is better.
Auto-generated definition — editorial description not yet written.
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.6 Sol (max) | 0.7 |
| 2 | GPT-5.5 (xhigh) | 0.6 |
| 3 | GPT-5.5 (high) | 0.6 |
| 4 | GPT-5.5 (medium) | 0.6 |
| 5 | Claude Sonnet 5 (Adaptive Reasoning, Xhigh Effort) | 0.6 |
| 6 | GLM-5.2 (max) | 0.5 |
| 7 | GPT-5.5 (low) | 0.5 |
| 8 | Claude Opus 4.7 | 0.5 |
| 9 | Kimi K2.6 | 0.5 |
| 10 | Kimi K2.5 (Reasoning) | 0.3 |