# STEMBench → AA-Omniscience Accuracy

Calibration · https://benchgap.net/calibration/1441 · data retrieved 2026-10-29

AA-Omniscience Accuracy is estimated from STEMBench with a inverse Michaelis–Menten curve fitted on 8 models measured on both: y = 0.26672·(x − 0.0000) / (0.0000 + 0.0404 − x), R² = 0.38, cross-validated error 13.3 pp. It is used for 0 estimates.

## Fitted on

| Model | STEMBench | AA-Omniscience Accuracy |
|---|---|---|
| [Claude Opus 5.5](https://benchgap.net/model/claude-opus-5-5.md) | 2.6% | 65.5% |
| [DeepSeek V4.1 Flash 1017](https://benchgap.net/model/deepseek-v4-1-flash.md) | 2.5% | 46.4% |
| [Gemini 4 Argon](https://benchgap.net/model/gemini-4-argon.md) | 2.4% | 48.3% |
| [GPT-6.1 Sol](https://benchgap.net/model/gpt-6-1-sol.md) | 2.8% | 61.8% |
| [GPT-6 Astra](https://benchgap.net/model/gpt-6-astra.md) | 2.8% | 61.6% |
| [MiMo-V2.6-Pro](https://benchgap.net/model/mimo-v2-6-pro.md) | 2.2% | 34.8% |
| [Mistral Large 4](https://benchgap.net/model/mistral-large-4.md) | 2.5% | 23.9% |
| [Muse Wonder](https://benchgap.net/model/muse-wonder.md) | 2.7% | 38.0% |