ScreenSpot-Pro — Different source cluster (Kimi-VL); tools condition unstated in name, kept separate to be safe.
| # | Model | Vendor | Best score | Runs | Last seen |
|---|---|---|---|---|---|
| 1 | Kimi VL A3B Thinking | Moonshot | 52.8 | 1 | 2026-06-05 |
| 2 | Qwen2.5 VL 72B | Alibaba | 43.6 | 1 | 2026-06-05 |
| 3 | Qwen2.5 VL 32B | Alibaba | 39.4 | 1 | 2026-06-05 |
| 4 | Kimi VL A3B Instruct | Moonshot | 35.4 | 2 | 2026-06-12 |
| 5 | GPT-4o | OpenAI | 0.8 | 2 | 2026-06-05 |
Best tracked score per model (default configuration; source-attributed and verification-tiered). Open a model for its full benchmark surface, provenance and pricing.