OpenVLM MMMU - Mechanical Engineering: leaderboard
Metric: Accuracy (%). Source: huggingface.co. 284 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5 | 70 |
| 2 | Gemini 2.5 Pro | 63.3 |
| 3 | GPT-5 Nano | 63.3 |
| 4 | Gemini 2.0 Flash | 63.3 |
| 5 | GPT-5 Mini (2025-08-07) | 63.3 |
| 6 | GPT-4.1 (2025-04-14) | 60 |
| 7 | GPT-4.5 | 60 |
| 8 | InternVL3-38B | 56.7 |
| 9 | Gemini 1.5 Pro (002) | 56.7 |
| 10 | Claude 3.7 Sonnet | 56.7 |
| 11 | Grok 2 (1212) | 56.7 |
| 12 | Qwen 2 VL 72B | 53.3 |
| 13 | Claude 3.5 Sonnet (20241022) | 53.3 |
| 14 | InternVL2.5-78B | 53.3 |
| 15 | InternVL3-8B | 50 |
Interactive version: theaggregate.ai/benchmark?slug=openvlm-mmmu-mechanical-engineering · How It Works · Data refreshed daily, snapshot 2026-09-05.