7B Medical AI Agent Beats o3 and GPT-5 by Learning Where and How to Look

7B打败o3、GPT-5！医学AI智能体让模型学会“看哪里、怎么看”

Shanghai Innovation Institute’s LeapQuest and three universities released Ophiuchus and MedScope, applying Think with Images and Think with Videos to medical AI; Ophiuchus-7B scored 68.0 on eight VQA benchmarks, above OpenAI-o3 at 62.2, Gemini 2.5 Pro at 61.8, and GPT-5 at 59.9.

Why it matters: HKR-H/K/R all pass: a 7B model beating o3/GPT-5 is a strong hook, 8 VQA benchmarks with 68.0 vs 62.2 add a testable claim, and medical specialist evaluation will trigger debate. Not a frontier-lab general model release, so it stays in 78–84.