# SWE-Marathon → Bug Hunt Bench

Calibration · https://benchgap.net/calibration/1488 · data retrieved 2026-10-29

Bug Hunt Bench is estimated from SWE-Marathon with a linear curve fitted on 5 models measured on both: y = 1.3895·x + -0.3865, R² = 0.94, cross-validated error 3.2 pp. It is used for 2 estimates.

| Estimated model | SWE-Marathon | Bug Hunt Bench | Source |
|---|---|---|---|
| [Step 5 Preview](https://benchgap.net/model/step-5-preview.md) | 72.7% | 62.4% | estimated ± 3.2 pp, low confidence |
| [Hy4 preview](https://benchgap.net/model/hy4-preview.md) | 31.9% | 5.7% | estimated ± 3.2 pp, low confidence |

## Fitted on

| Model | SWE-Marathon | Bug Hunt Bench |
|---|---|---|
| [GLM-5.3](https://benchgap.net/model/glm-5-3.md) | 42.5% | 19.0% |
| [Kimi K3.1 Pro](https://benchgap.net/model/kimi-3-1-pro.md) | 51.7% | 36.0% |
| [Kimi K3](https://benchgap.net/model/kimi-k3.md) | 42.0% | 21.0% |
| [Muse Wonder](https://benchgap.net/model/muse-wonder.md) | 53.1% | 31.7% |
| [Muse Wonder Full](https://benchgap.net/model/muse-wonder-full.md) | 57.5% | 42.0% |