| # | Model | Vendor | Best score | Runs | Last seen |
|---|---|---|---|---|---|
| 1 | Phi 3.5 Vision Instruct | Microsoft | 0.7 | 1 | 2026-08-12 |
| 2 | North-Micro-Vision-Instruct | Cohere | 0.6 | 1 | 2026-08-12 |
| 3 | Qwen3 VL 2B Instruct | Alibaba | 0.6 | 1 | 2026-08-12 |
| 4 | Ministral 3 3B | Mistral | 0.5 | 1 | 2026-08-12 |
| 5 | Qwen3.5 2B | Alibaba | 0.5 | 1 | 2026-08-12 |
| 6 | LFM2.5-VL-1.6B | Liquid AI | 0.4 | 1 | 2026-08-12 |
| 7 | Gemma 4 E2B | 0.4 | 1 | 2026-08-12 |
Best tracked score per model (default configuration; source-attributed and verification-tiered). Open a model for its full benchmark surface, provenance and pricing.