OpenVLM MMMU - Energy and Power: leaderboard
OpenCompass OpenVLM MMMU validation subcategory score for visual college-level questions in energy and power.
Metric: Accuracy (%). Source: huggingface.co. Saturation forecast: Around January 2027. 284 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5 | 80 | #98 |
| 2 | Gemini 2.0 Flash | 73.3 | #342 |
| 3 | Gemma 3 27B | 73.3 | #607 |
| 4 | GPT-5 Mini (2025-08-07) | 73.3 | #172 |
| 5 | GPT-5 Nano | 70 | #425 |
| 6 | Gemini 2.5 Pro | 66.7 | #152 |
| 7 | InternVL3-14B | 66.7 | #505 |
| 8 | GPT-4o ChatGPT | 66.7 | |
| 9 | InternVL3-8B | 63.3 | #620 |
| 10 | InternVL3-38B | 63.3 | #405 |
| 11 | InternVL3-78B | 60 | #352 |
| 12 | Claude 3.7 Sonnet | 60 | #249 |
| 13 | Claude 3.5 Sonnet | 60 | #350 |
| 14 | Claude 3.5 Sonnet (20241022) | 56.7 | #291 |
| 15 | GPT-4.1 Nano | 56.7 | #731 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=openvlm-mmmu-energy-and-power · How It Works · Data refreshed daily, snapshot 2026-10-20.