# OSWorld-Verified → DeepSearchQA

Calibration · https://benchgap.net/calibration/osworld-verified/current/to/deepsearchqa/current · data retrieved 2026-11-18

DeepSearchQA is estimated from OSWorld-Verified with a Hill curve fitted on 7 models measured on both: y = 0.6926 + (1.2000 − 0.6926)·x^6.00 / (0.89987^6.00 + x^6.00), R² = 0.38, cross-validated error 7.9 pp. It is used for 10 estimates.

| Estimated model | OSWorld-Verified | DeepSearchQA | Source |
|---|---|---|---|
| [Qwen3.8 Max](https://benchgap.net/model/qwen3-8-max.md) | 86.1% | 91.3% | estimated ± 7.9 pp, low confidence |
| [Holo3-35B-A3B](https://benchgap.net/model/holo3-35b-a3b.md) | 82.6% | 88.2% | estimated ± 7.9 pp, low confidence |
| [MiMo-V2.6-Pro](https://benchgap.net/model/mimo-v2-6-pro.md) | 82.0% | 87.7% | estimated ± 7.9 pp, low confidence |
| [MiMo-V2.6-Flash](https://benchgap.net/model/mimo-v2-6-flash.md) | 80.8% | 86.7% | estimated ± 7.9 pp, low confidence |
| [Holo3-122B-A10B](https://benchgap.net/model/holo3-122b-a10b.md) | 78.9% | 85.1% | estimated ± 7.9 pp, low confidence |
| [UI-Mate-27B](https://benchgap.net/model/ui-mate-27b.md) | 77.0% | 83.6% | estimated ± 7.9 pp, low confidence |
| [Claude Opus 4.5](https://benchgap.net/model/claude-opus-4-5.md) | 66.3% | 76.3% | estimated ± 7.9 pp, low confidence |
| [UI-Mate-9B](https://benchgap.net/model/ui-mate-9b.md) | 66.2% | 76.2% | estimated ± 7.9 pp, low confidence |
| [GPT-5.3 Codex](https://benchgap.net/model/gpt-5-3-codex.md) | 64.7% | 75.4% | estimated ± 7.9 pp, low confidence |
| [Claude Sonnet 4.5](https://benchgap.net/model/claude-sonnet-4-5.md) | 61.4% | 73.9% | estimated ± 7.9 pp, low confidence |

## Fitted on

| Model | OSWorld-Verified | DeepSearchQA |
|---|---|---|
| [Claude Opus 4.6](https://benchgap.net/model/claude-opus-4-6.md) | 72.7% | 73.7% |
| [Claude Opus 4.8](https://benchgap.net/model/claude-opus-4-8.md) | 83.4% | 93.1% |
| [GPT-5.4](https://benchgap.net/model/gpt-5-4.md) | 75.0% | 73.6% |
| [Kimi K2.6](https://benchgap.net/model/kimi-2-6.md) | 73.1% | 92.5% |
| [Kimi K2.5](https://benchgap.net/model/kimi-k2-5.md) | 63.3% | 77.1% |
| [Muse Glimmer 30B](https://benchgap.net/model/muse-glimmer-30b.md) | 65.9% | 74.6% |
| [Muse Spark 1.1](https://benchgap.net/model/muse-spark-1-1.md) | 80.8% | 84.9% |
