OpenVLM MMT-Bench - Emo — leaderboard
Metric: Score (%). Source: huggingface.co. 207 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | InternVL3-78B | 66.7 |
| 2 | InternVL3-38B | 61.7 |
| 3 | InternVL3-8B | 60.8 |
| 4 | GPT-4.1 (2025-04-14) | 60.8 |
| 5 | InternVL3-14B | 57.5 |
| 6 | GPT-4.1 Mini | 56.7 |
| 7 | Qwen 2 VL 72B | 55.8 |
| 8 | Grok 2 (1212) | 55.8 |
| 9 | Gemini 1.5 Pro | 55 |
| 10 | MiniCPM-V-2.6 | 54.2 |
| 11 | Qwen 2 VL 7B | 53.3 |
| 12 | Qwen 2 VL 2B | 53.3 |
| 13 | Aquila-VL-2B | 53.3 |
| 14 | InternVL2-8B | 52.5 |
| 15 | Ovis2-8B | 51.7 |
Interactive version: theaggregate.ai/benchmark?slug=openvlm-mmt-bench-emo · How the rankings work · Data refreshed daily, snapshot 2026-07-22.