OpenVLM MMMU - Humanities and Social Science — leaderboard

Metric: Accuracy (%). Source: huggingface.co. 284 models tracked.

Top models

#ModelScore
1Gemini 2.5 Pro85
2GPT-5 (2025-08-07)85
3GPT-4o ChatGPT82.5
4GPT-4.1 (2025-04-14)81.7
5InternVL2.5-78B80.8
6GPT-5 Mini (2025-08-07)80.8
7InternVL3-78B80
8Qwen 2 VL 72B80
9Grok 2 (1212)80
10GPT-4.580
11InternVL3-38B79.2
12InternVL3-14B79.2
13Gemma 3 27B79.2
14Gemini 1.5 Flash (002)78.3
15InternVL3-8B77.5

Interactive version: theaggregate.ai/benchmark?slug=openvlm-mmmu-humanities-and-social-science · How the rankings work · Data refreshed daily, snapshot 2026-07-22.