| # | Model | Vendor | Best score | Runs | Last seen |
|---|---|---|---|---|---|
| 1 | LFM2.5-VL-3B | Liquid AI | 83 | 2 | 2026-08-14 |
| 2 | LFM2-VL-3B (3.1B) | Liquid AI | 81.9 | 1 | 2026-08-12 |
| 3 | LFM2-VL-3B | Liquid AI | 81.9 | 1 | 2026-08-22 |
| 4 | Gemma 4 E2B | 75.7 | 2 | 2026-08-12 | |
| 5 | LFM2.5-VL-450M-Extract | Liquid AI | 68.1 | 1 | 2026-08-09 |
| 6 | LFM2-VL-450M | Liquid AI | 54.3 | 1 | 2026-08-09 |
| 7 | Qwen3.5 2B | Alibaba | 0.7 | 1 | 2026-08-12 |
| 8 | Ministral 3 3B | Mistral | 0.7 | 1 | 2026-08-12 |
| 9 | North-Micro-Vision-Instruct | Cohere | 0.7 | 1 | 2026-08-12 |
| 10 | Qwen3 VL 2B Instruct | Alibaba | 0.7 | 1 | 2026-08-12 |
| 11 | LFM2.5-VL-1.6B | Liquid AI | 0.7 | 1 | 2026-08-12 |
| 12 | Phi 3.5 Vision Instruct | Microsoft | 0.7 | 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.