| # | Model | Vendor | Best score | Runs | Last seen |
|---|---|---|---|---|---|
| 1 | Mage-VL-4B | Microsoft | 66.5 | 1 | 2026-07-26 |
| 2 | Qwen3 VL 4B (Reasoning) | Alibaba | 63.2 | 1 | 2026-07-26 |
| 3 | Qwen2.5 Omni 7B | Alibaba | 61.6 | 1 | 2026-08-23 |
| 4 | Phi 4 R V 15B | Microsoft | 57.8 | 1 | 2026-07-26 |
| 5 | Phi 4 MM 5.6B | Microsoft | 32.5 | 1 | 2026-07-26 |
Best tracked score per model (default configuration; source-attributed and verification-tiered). Open a model for its full benchmark surface, provenance and pricing.