Skip to content

medgemma 1.5 4b it

google/medgemma-1.5-4b-it
Open Source · chat · open-weights
Open Alert me on changes
Context
Max output
Weights
Open
API $/1M
Modalities
text · image
Released
07 Jan 2026
License: other · google/medgemma-1.5-4b-it
Download image Share on X Share on LinkedIn
AI summary
● machine-written

MedGemma 1.5 4B IT: Medical-trained multimodal chat model

MedGemma 1.5 4B IT is an open-weights model based on Gemma 3, further trained on medical images and text including chest X-rays, histopathology, dermatology, and electronic health records. It supports both text and image inputs for medical contexts. The model is distributed under an unspecified "other" license via Hugging Face.

What's new
  • Based on Gemma 3 with additional medical image and text training
  • Trained on diverse medical data: chest X-rays, histopathology, dermatology, fundus images, CT, MR, and EHR
  • Supports multimodal input: text and image comprehension
  • Uses SigLIP image encoder pre-trained on de-identified medical data
Best for
Medical image analysis and interpretationClinical reasoning from radiology and pathology imagesMedical document and EHR text processingDermatology and ophthalmology image understanding
Sources

Source: https://huggingface.co/google/medgemma-1.5-4b-it