# We-Math → CountBench

Calibration · https://benchgap.net/calibration/1668 · data retrieved 2026-10-29

CountBench is estimated from We-Math with a offset logistic curve fitted on 6 models measured on both: y = 0.9183 + (0.9730 − 0.9183) / (1 + exp(−200.00·(x − 0.8487))), R² = 0.92, cross-validated error 1.8 pp. It is used for 0 estimates.

## Fitted on

| Model | We-Math | CountBench |
|---|---|---|
| [Claude Opus 4.5](https://benchgap.net/model/claude-opus-4-5.md) | 70.0% | 90.6% |
| [Gemini 3 Pro](https://benchgap.net/model/gemini-3-pro.md) | 86.9% | 97.3% |
| [GPT-5.2](https://benchgap.net/model/gpt-5-2.md) | 79.0% | 91.9% |
| [Kimi K2.5](https://benchgap.net/model/kimi-k2-5.md) | 84.7% | 94.1% |
| [LFM3-VL-2B](https://benchgap.net/model/lfm3-vl-2b.md) | 68.6% | 93.0% |
| [Qwen3.5 397B](https://benchgap.net/model/qwen3-5-397b.md) | 87.9% | 97.2% |