# PhysicianBench → HealthBench (raw, Artificial Analysis)

Calibration · https://benchgap.net/calibration/physicianbench/current/to/aa-health-bench/current · data retrieved 2026-11-21

HealthBench (raw, Artificial Analysis) is estimated from PhysicianBench with a Michaelis–Menten + offset curve fitted on 5 models measured on both: y = 0.0000 + 0.8920·x / (0.26551 + x), R² = 0.47, cross-validated error 4.3 pp. It is used for 0 estimates.

## Fitted on

| Model | PhysicianBench | HealthBench (raw, Artificial Analysis) |
|---|---|---|
| [Claude Fable 5.1](https://benchgap.net/model/claude-fable-5-1.md) | 61.0% | 63.9% |
| [Claude Fable 5.2](https://benchgap.net/model/claude-fable-5-2.md) | 64.4% | 58.8% |
| [Claude Opus 5.5](https://benchgap.net/model/claude-opus-5-5.md) | 68.4% | 64.3% |
| [Claude Sonnet 5.5](https://benchgap.net/model/claude-sonnet-5-5.md) | 63.2% | 65.9% |
| [GPT-6 Astra](https://benchgap.net/model/gpt-6-astra.md) | 48.0% | 56.9% |
