AnalysisAI ModelsOctober 6, 2026

MedGemma medical vision-language models built on Gemma 3

Read original source →nature.com

MedGemma, a collection of medical vision-language foundation models based on Gemma 3, improves out-of-distribution performance by 2.6–10% on medical image QA, 15.5–18.1% on chest X-ray finding classification and 10.8% on agentic evaluations versus base models. The collection includes MedSigLIP, a medically tuned vision encoder derived from SigLIP.

1 source

More stories today

Open the live feed