# Toolathlon-Verified → MazeBench

Calibration · https://benchgap.net/calibration/1536 · data retrieved 2026-10-29

MazeBench is estimated from Toolathlon-Verified with a linear curve fitted on 10 models measured on both: y = 1.2724·x + -0.9156, R² = 0.61, cross-validated error 4.0 pp. It is used for 5 estimates.

| Estimated model | Toolathlon-Verified | MazeBench | Source |
|---|---|---|---|
| [DeepSeek V4 Pro 0813](https://benchgap.net/model/deepseek-v4-pro-0813.md) | 74.1% | 2.7% | estimated ± 4.0 pp, high confidence |
| [Hy4 preview](https://benchgap.net/model/hy4-preview.md) | 74.1% | 2.7% | estimated ± 4.0 pp, high confidence |
| [Step 5 Preview](https://benchgap.net/model/step-5-preview.md) | 74.1% | 2.7% | estimated ± 4.0 pp, high confidence |
| [Qwen3.8-Flash-Next](https://benchgap.net/model/qwen3-8-flash-next.md) | 73.5% | 2.0% | estimated ± 4.0 pp, high confidence |
| [Ornith-1.5-397B](https://benchgap.net/model/ornith-1-5-397b.md) | 71.2% | 0.0% | estimated ± 4.0 pp, high confidence |

## Fitted on

| Model | Toolathlon-Verified | MazeBench |
|---|---|---|
| [Claude Fable 5.1](https://benchgap.net/model/claude-fable-5-1.md) | 77.8% | 10.0% |
| [Claude Opus 5](https://benchgap.net/model/claude-opus-5.md) | 80.6% | 12.0% |
| [Claude Opus 5.5](https://benchgap.net/model/claude-opus-5-5.md) | 77.8% | 13.0% |
| [Claude Sonnet 5.5](https://benchgap.net/model/claude-sonnet-5-5.md) | 77.8% | 8.0% |
| [DeepSeek V4 Flash 0731](https://benchgap.net/model/deepseek-v4-flash-0731.md) | 70.3% | 0.0% |
| [GLM-5.3](https://benchgap.net/model/glm-5-3.md) | 73.0% | 0.0% |
| [GLM-5.3-Flash](https://benchgap.net/model/glm-5-3-flash.md) | 78.4% | 0.0% |
| [Kimi K3](https://benchgap.net/model/kimi-k3.md) | 73.2% | 1.0% |
| [Muse Spark Open](https://benchgap.net/model/muse-spark-open.md) | 72.8% | 0.0% |
| [Qwen3.8 Max](https://benchgap.net/model/qwen3-8-max.md) | 72.5% | 0.0% |