# Terminal-Bench-Science 0.1 → Terminal-Bench 4.0

Calibration · https://benchgap.net/calibration/terminal-bench-science/current/to/terminal-bench/current · data retrieved 2026-11-21

Terminal-Bench 4.0 is estimated from Terminal-Bench-Science 0.1 with a Michaelis–Menten curve fitted on 11 models measured on both: y = 0.7823·x / (0.14226 + x), R² = 0.89, cross-validated error 6.9 pp. It is used for 0 estimates.

## Fitted on

| Model | Terminal-Bench-Science 0.1 | Terminal-Bench 4.0 |
|---|---|---|
| [Claude Fable 5.1](https://benchgap.net/model/claude-fable-5-1.md) | 52.6% | 55.8% |
| [Claude Fable 5.2](https://benchgap.net/model/claude-fable-5-2.md) | 74.6% | 67.6% |
| [Claude Mythos 5.2](https://benchgap.net/model/claude-mythos-5-2.md) | 75.3% | 70.5% |
| [Claude Opus 5.5](https://benchgap.net/model/claude-opus-5-5.md) | 58.7% | 66.4% |
| [Claude Sonnet 5.5](https://benchgap.net/model/claude-sonnet-5-5.md) | 59.9% | 70.6% |
| [ERNIE 5.5 Flash Preview](https://benchgap.net/model/ernie-5-5-flash-preview.md) | 21.0% | 46.5% |
| [GPT-6.1 Terra](https://benchgap.net/model/gpt-6-1-terra.md) | 58.3% | 50.3% |
| [GPT-6 Astra](https://benchgap.net/model/gpt-6-astra.md) | 64.6% | 57.9% |
| [Hy4](https://benchgap.net/model/hy4.md) | 19.8% | 44.4% |
| [Inkling 2 Mini](https://benchgap.net/model/inkling-2-mini.md) | 1.4% | 5.6% |
| [Kimi K3.1 Pro](https://benchgap.net/model/kimi-3-1-pro.md) | 38.1% | 65.7% |
