Official benchmarks map / ScreenSpot-Pro
beipo-ai/NorthCUA-Pro is ranked first on the ScreenSpot-Pro leaderboard with 93.3, 1.0% ahead of beipo-ai/NorthCUA-Mini. Read live from the Hugging Face API on November 6, 2026.
| # | Model | Organization | Score |
|---|---|---|---|
| 1 | beipo-ai/NorthCUA-Pro | beipo-ai | 93.3 |
| 2 | beipo-ai/NorthCUA-Mini | beipo-ai | 92.4 |
| 3 | meta-models/Muse-Glimmer-30B | meta-models | 75.4 |
| 4 | beipo-ai/NorthCUA-Nano | beipo-ai | 72.9 |
| 5 | Qwen/Qwen3.5-122B-A10B | Qwen | 70.4 |
| 6 | Qwen/Qwen3.5-27B | Qwen | 70.3 |
| 7 | inclusionAI/UI-Venus-1.5-8B | inclusionAI | 68.4 |
| 8 | Qwen/Qwen3.5-397B-A17B | Qwen | 65.6 |
| 9 | Salesforce/GTA1-32B | Salesforce | 63.6 |
| 10 | Qwen/Qwen3-VL-235B-A22B-Instruct | Qwen | 62 |
ScreenSpot-Pro is a dataset that Hugging Face marks as an official benchmark. Its leaderboard is built automatically from the .eval_results files that model repositories publish, so entries are mostly self-reported by model authors and evaluation settings differ between submissions. A verified tag appears only where Hugging Face marks an entry as verified.
Original board: huggingface.co/datasets/likaixin/ScreenSpot-Pro
Part of the Hugging Face official benchmarks map by QuantID. Unofficial. Not affiliated with or endorsed by Hugging Face.