OpenVLM MMStar - Fine-Grained Perception — leaderboard
Metric: Accuracy (%). Source: huggingface.co. 284 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 2.5 Pro | 66.8 |
| 2 | InternVL3-38B | 64.4 |
| 3 | InternVL3-78B | 64 |
| 4 | Qwen 2 VL 72B | 61.2 |
| 5 | InternVL3-8B | 60 |
| 6 | InternVL3-14B | 60 |
| 7 | GPT-5 (2025-08-07) | 58.8 |
| 8 | GPT-5 Mini (2025-08-07) | 58.8 |
| 9 | GPT-4.1 Mini | 58.4 |
| 10 | GPT-4.5 | 58.4 |
| 11 | Gemini 2.0 Flash | 57.6 |
| 12 | Gemini 1.5 Pro (002) | 57.2 |
| 13 | GPT-4o ChatGPT | 56.8 |
| 14 | Qwen 2 VL 7B | 56.4 |
| 15 | Ovis2-8B | 56.4 |
Interactive version: theaggregate.ai/benchmark?slug=openvlm-mmstar-fine-grained-perception · How the rankings work · Data refreshed daily, snapshot 2026-07-22.