OpenVLM MMMU - Psychology — leaderboard
Metric: Accuracy (%). Source: huggingface.co. 284 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | InternVL2.5-78B | 86.7 |
| 2 | Gemini 2.5 Pro | 83.3 |
| 3 | GPT-5 Nano | 83.3 |
| 4 | InternVL3-38B | 80 |
| 5 | Grok 2 (1212) | 80 |
| 6 | GPT-4.1 (2025-04-14) | 80 |
| 7 | GPT-4o ChatGPT | 80 |
| 8 | GPT-5 Mini (2025-08-07) | 80 |
| 9 | Qwen 2 VL 72B | 76.7 |
| 10 | InternVL3-14B | 76.7 |
| 11 | Gemini 1.5 Pro (002) | 76.7 |
| 12 | Gemini 1.5 Flash (002) | 76.7 |
| 13 | Gemini 1.5 Pro | 76.7 |
| 14 | GPT-5 (2025-08-07) | 76.7 |
| 15 | InternVL3-78B | 73.3 |
Interactive version: theaggregate.ai/benchmark?slug=openvlm-mmmu-psychology · How the rankings work · Data refreshed daily, snapshot 2026-07-22.