# Video-MME (with subtitle) → SimpleVQA

Calibration · https://benchgap.net/calibration/1635 · data retrieved 2026-10-29

SimpleVQA is estimated from Video-MME (with subtitle) with a inverse Michaelis–Menten curve fitted on 10 models measured on both: y = 1.60743·(x − 0.2809) / (0.2809 + 2.0000 − x), R² = 0.54, cross-validated error 10.8 pp. It is used for 0 estimates.

## Fitted on

| Model | Video-MME (with subtitle) | SimpleVQA |
|---|---|---|
| [Claude Opus 4.5](https://benchgap.net/model/claude-opus-4-5.md) | 77.6% | 65.7% |
| [Gemini 3 Pro](https://benchgap.net/model/gemini-3-pro.md) | 88.4% | 73.2% |
| [GPT-5.2](https://benchgap.net/model/gpt-5-2.md) | 86.0% | 55.8% |
| [Kimi K2.5](https://benchgap.net/model/kimi-k2-5.md) | 87.4% | 71.2% |
| [LFM3-VL-2B](https://benchgap.net/model/lfm3-vl-2b.md) | 69.0% | 36.3% |
| [Qwen3.5 397B](https://benchgap.net/model/qwen3-5-397b.md) | 87.5% | 67.1% |
| [Qwen3.6-27B](https://benchgap.net/model/qwen3-6-27b.md) | 87.7% | 56.1% |
| [Qwen3.6-35B-A3B](https://benchgap.net/model/qwen3-6-35b-a3b.md) | 86.6% | 58.9% |
| [Qwen3.7 Plus](https://benchgap.net/model/qwen3-7-plus.md) | 88.0% | 81.7% |
| [Qwen3.8 Max](https://benchgap.net/model/qwen3-8-max.md) | 90.4% | 75.0% |