MMBU - Ungrounded Classification (Open-Ended): leaderboard

Metric: Open-ended (free-form) ungrounded classification of the whole image: micro-averaged F1 x100 (0-100) over MMBU's biomedical images from 35 submodalities (95% bootstrap intervals not published); zero-shot, official checkpoints; Qwen3-32B judges strict biomedical equivalence. Source: arxiv.org. Saturation forecast: Around August 2028. 17 models tracked.

Top models

#ModelScore
1Qwen 3 VL 32B Instruct13.7
2InternVL3.5-8B11.1
3GPT-5.4 Mini10.4
4Qwen 2.5 VL 32B Instruct8.8
5Qwen 3 VL 8B Instruct8.6
6Qwen 3 VL 4B Instruct7.9
7Lingshu-32B7.3
8GPT-4.1 Mini7
9Lingshu-7B5.4
10MedGemma-4B-IT4.3
11Qwen 2.5 VL 7B Instruct3.1
12Gemma 3 4B (IT)2.2

Interactive version: theaggregate.ai/benchmark?slug=mmbu-ungrounded-classification-open-ended · How It Works · Data refreshed daily, snapshot 2026-09-29.