# WideResearch → Toolathlon

Calibration · https://benchgap.net/calibration/1690 · data retrieved 2026-10-29

Toolathlon is estimated from WideResearch with a Hill curve fitted on 7 models measured on both: y = 0.1948 + (1.2000 − 0.1948)·x^6.00 / (0.94486^6.00 + x^6.00), R² = 0.71, cross-validated error 5.2 pp. It is used for 2 estimates.

| Estimated model | WideResearch | Toolathlon | Source |
|---|---|---|---|
| [Hy4 preview](https://benchgap.net/model/hy4-preview.md) | 83.9% | 52.5% | estimated ± 5.2 pp, low confidence |
| [Qwen3.8 Max](https://benchgap.net/model/qwen3-8-max.md) | 81.9% | 49.4% | estimated ± 5.2 pp, low confidence |

## Fitted on

| Model | WideResearch | Toolathlon |
|---|---|---|
| [Claude Opus 4.5](https://benchgap.net/model/claude-opus-4-5.md) | 76.4% | 43.5% |
| [GLM-5](https://benchgap.net/model/glm-5.md) | 69.8% | 38.0% |
| [Kimi K2.6](https://benchgap.net/model/kimi-2-6.md) | 80.8% | 50.0% |
| [Kimi K2.5](https://benchgap.net/model/kimi-k2-5.md) | 72.7% | 27.8% |
| [Qwen3.5 397B](https://benchgap.net/model/qwen3-5-397b.md) | 74.0% | 36.3% |
| [Qwen3.6-35B-A3B](https://benchgap.net/model/qwen3-6-35b-a3b.md) | 60.1% | 26.9% |
| [Qwen3.6 Plus](https://benchgap.net/model/qwen3-6-plus.md) | 74.3% | 39.8% |