Calibration
FrontierCode 2.0 → CursorBench 3.2
CursorBench 3.2 is estimated from FrontierCode 2.0 with a Hill curve fitted on 6 models measured on both: y = 0.5996 + (0.8690 − 0.5996)·x^6.00 / (0.33655^6.00 + x^6.00), R² = 0.79, cross-validated error 4.1 pp. It is used for 6 estimates.
| Estimated model | FrontierCode 2.0 | CursorBench 3.2 | Source |
|---|---|---|---|
| GPT-6.2 Astra | 44.2% | 82.5% | estimated ± 4.1 pp, low confidence |
| GPT-6.2 Sol | 43.8% | 82.3% | estimated ± 4.1 pp, low confidence |
| Inkling 2 Small | 22.6% | 62.2% | estimated ± 4.1 pp, medium confidence |
| Kimi K3.1 Pro | 19.9% | 61.1% | estimated ± 4.1 pp, medium confidence |
| Mistral Large 4 | 19.7% | 61.0% | estimated ± 4.1 pp, medium confidence |
| Muse Spark Open | 6.5% | 60.0% | estimated ± 4.1 pp, low confidence |
Fitted on
| Model | FrontierCode 2.0 | CursorBench 3.2 |
|---|---|---|
| Claude Fable 5.1 | 34.0% | 73.4% |
| Claude Opus 4.8 | 15.3% | 62.3% |
| Claude Opus 5 | 28.1% | 70.0% |
| GPT-5.5 | 23.6% | 58.4% |
| GPT-5.6 Sol | 28.9% | 67.2% |
| Kimi K3 | 18.9% | 60.8% |