LingSante-3.1-air-VL tops medical imaging evals across derm, labs and X-rays

Dr_Alex_Crimi · x · 2026-11-23

Dr. Alex Crimi shares a comparison of US vs Chinese, closed vs open models on medical image interpretation (Breakthrough benchmarks): inclusionAI's open-weights LingSante-3.1-air-VL ranks first on all three tasks — skin lesions (AFUMED-Derm, 70.01), lab reports (AFUMED-Report, 90.70), and chest X-rays (CheXpert Plus, 46.21) — beating Gemini 3.8 Flash, GPT-6.1 Sol and Claude Opus 5.5. Other Chinese models on the board include DeepSeek-V4.1-Flash-1017, GLM-5.3-Flash, Kimi-K3.1-Pro and MiniMax-M3.1-Flash.

Original post →

More from Models

Models channel →