OpenVLM MMT-Bench - RR: leaderboard
Metric: Score (%). Source: huggingface.co. 207 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Grok 2 (1212) | 82.5 |
| 2 | GPT-4.1 (2025-04-14) | 78.8 |
| 3 | GPT-4.1 Mini | 77.5 |
| 4 | InternVL3-78B | 76.2 |
| 5 | InternVL3-38B | 75 |
| 6 | Gemini 1.5 Flash | 75 |
| 7 | Qwen 2 VL 72B | 73.8 |
| 8 | InternVL3-8B | 72.5 |
| 9 | Claude 3.7 Sonnet | 71.2 |
| 10 | Gemini 1.5 Pro | 71.2 |
| 11 | Qwen 2 VL 7B | 70 |
| 12 | Aquila-VL-2B | 70 |
| 13 | Gemma 3 27B | 70 |
| 14 | Qwen 2 VL 2B | 68.8 |
| 15 | Claude 3.5 Sonnet | 66.2 |
Interactive version: theaggregate.ai/benchmark?slug=openvlm-mmt-bench-rr · How It Works · Data refreshed daily, snapshot 2026-09-05.