# CountBench → DynaMath

Calibration · https://benchgap.net/calibration/countbench/current/to/dynamath/current · data retrieved 2026-11-21

DynaMath is estimated from CountBench with a Michaelis–Menten curve fitted on 7 models measured on both: y = 2.0000·x / (1.30876 + x), R² = 0.32, cross-validated error 2.9 pp. It is used for 0 estimates.

## Fitted on

| Model | CountBench | DynaMath |
|---|---|---|
| [Claude Opus 4.5](https://benchgap.net/model/claude-opus-4-5.md) | 90.6% | 79.7% |
| [Gemini 3 Pro](https://benchgap.net/model/gemini-3-pro.md) | 97.3% | 85.1% |
| [GPT-5.2](https://benchgap.net/model/gpt-5-2.md) | 91.9% | 86.8% |
| [Kimi K2.5](https://benchgap.net/model/kimi-k2-5.md) | 94.1% | 84.4% |
| [LFM3-VL-2B](https://benchgap.net/model/lfm3-vl-2b.md) | 93.0% | 79.2% |
| [Qwen3.5 397B](https://benchgap.net/model/qwen3-5-397b.md) | 97.2% | 86.3% |
| [Qwen3.6-27B](https://benchgap.net/model/qwen3-6-27b.md) | 97.8% | 85.6% |
