OSWorld-G - Text Matching: leaderboard
OSWorld-G samples whose target is located by its visible text (268 of 564); accuracy of the model's click.
Metric: Accuracy (%). Source: arxiv.org. Saturation forecast: Estimated already saturated. 6 models tracked. Updated 17 Nov 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Jedi 3B-1080p | 67.4 | |
| 2 | Jedi-7B-1080p | 65.9 | |
| 3 | Qwen 2.5 VL 32B Instruct | 63.2 | #1074 |
| 4 | UGround-V1-7B | 51.3 | |
| 5 | Qwen 2.5 VL 7B Instruct | 45.6 | #1893 |
| 6 | Qwen 2.5 VL 3B Instruct | 41.4 | #2648 |
No result here: #1 Claude Fable 5.2, #2 GPT-6.2 Astra, #3 Gemini 4 Carbon.
Interactive version: theaggregate.ai/benchmark?slug=osworld-g-text-matching · How It Works · Data refreshed daily, snapshot 2026-11-17.