AutoMedBench - Visual Question Answering: leaderboard

Metric: Overall run score (0-100) on the visual question answering track (5 radiology, pathology and video VQA tasks): the mean of the held-out task score (VQA accuracy, scaled to 0-100) and the workflow agentic score (Plan, Setup, Validate, Inference and Submit stages weighted 25/15/35/15/10, LLM judge for the first three); each model is the single agent of the benchmark harness (shared code-execution interface, system prompt, tool schema and default decoding; no vendor agent framework), runs averaged over the Lite and Standard tiers; higher is better. Source: arxiv.org. Saturation forecast: Around 2029. 6 models tracked.

Top models

#ModelScore
1GLM-564
2Gemini 3.1 Pro (Preview)62.3
3Qwen 3.5 397B A17B57
4MiniMax-M2.555.8
5Claude Opus 4.655.5
6GPT-5.436.5

Interactive version: theaggregate.ai/benchmark?slug=automedbench-visual-question-answering · How It Works · Data refreshed daily, snapshot 2026-09-29.