OpenVLM MMT-Bench - Traffic Participants Understanding — leaderboard
Metric: Score (%). Source: huggingface.co. 207 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Ovis2-8B | 75 |
| 2 | InternVL3-8B | 65 |
| 3 | GPT-4.1 Nano | 65 |
| 4 | Gemma 3 27B | 65 |
| 5 | InternVL2-8B | 60 |
| 6 | Qwen 2 VL 7B | 60 |
| 7 | InternVL3-38B | 60 |
| 8 | Gemini 1.5 Pro | 60 |
| 9 | Gemini 1.5 Flash | 60 |
| 10 | GPT-4.1 Mini | 55 |
| 11 | InternVL3-78B | 55 |
| 12 | Qwen 2 VL 2B | 55 |
| 13 | Aquila-VL-2B | 55 |
| 14 | MiniCPM-V-2.6 | 55 |
| 15 | Pixtral-12B | 55 |
Interactive version: theaggregate.ai/benchmark?slug=openvlm-mmt-bench-traffic-participants-understanding · How the rankings work · Data refreshed daily, snapshot 2026-07-22.