# ERQA → OmniDocBench 1.5

Calibration · https://benchgap.net/calibration/erqa/current/to/omnidocbench15/current · data retrieved 2026-11-18

OmniDocBench 1.5 is estimated from ERQA with a linear curve fitted on 7 models measured on both: y = 0.1604·x + 0.7913, R² = 0.45, cross-validated error 2.5 pp. It is used for 2 estimates.

| Estimated model | ERQA | OmniDocBench 1.5 | Source |
|---|---|---|---|
| [Glimpse](https://benchgap.net/model/glimpse.md) | 83.8% | 92.6% | estimated ± 2.5 pp, low confidence |
| [Claude Opus 4.6](https://benchgap.net/model/claude-opus-4-6.md) | 51.6% | 87.4% | estimated ± 2.5 pp, low confidence |

## Fitted on

| Model | ERQA | OmniDocBench 1.5 |
|---|---|---|
| [Claude Opus 4.5](https://benchgap.net/model/claude-opus-4-5.md) | 46.8% | 87.7% |
| [Gemini 3 Pro](https://benchgap.net/model/gemini-3-pro.md) | 70.5% | 88.5% |
| [GPT-5.2](https://benchgap.net/model/gpt-5-2.md) | 59.8% | 85.7% |
| [Qwen3.5 397B](https://benchgap.net/model/qwen3-5-397b.md) | 67.5% | 90.8% |
| [Qwen3.7 Plus](https://benchgap.net/model/qwen3-7-plus.md) | 69.8% | 91.4% |
| [Qwen3.8-27B](https://benchgap.net/model/qwen3-8-27b.md) | 65.5% | 91.1% |
| [Qwen3.8 Max](https://benchgap.net/model/qwen3-8-max.md) | 77.8% | 92.1% |
