OpenVLM MMMU - Public Health — leaderboard

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

Top models

#ModelScore
1Claude 3.5 Sonnet (20241022)96.7
2Claude 3.7 Sonnet96.7
3Claude 3.5 Sonnet96.7
4InternVL3-78B93.3
5Gemini 1.5 Pro (002)93.3
6InternVL2.5-78B93.3
7GPT-4.1 (2025-04-14)93.3
8GPT-4.593.3
9GPT-5 (2025-08-07)93.3
10InternVL3-38B90
11InternVL3-14B90
12Gemini 2.0 Flash90
13GPT-4o ChatGPT90
14Gemini 1.5 Pro86.7
15GPT-5 Mini (2025-08-07)86.7

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