MedGemma 1.5

4B-parameter medical multimodal model built on Gemma 3, trained for medical text and image comprehension. MedGemma 1.5 expands to high-dimensional medical imaging: CT/MRI volumes (3D), whole-slide histopathology, anatomical bounding-box localization on radiographs, and multi-timepoint chest X-ray analysis with prior image context. Also improves clinical document understanding (lab reports, EHRs).

Over MedGemma 1.0: +11% 3D MRI classification, +47% macro F1 on whole-slide pathology, +35% chest X-ray localization IoU, +22% EHR QA accuracy, +18% lab report extraction F1. Part of Google's Health AI Developer Foundations. Open weights on HuggingFace (terms of use required).