unsloth/medgemma-4b-it

The unsloth/medgemma-4b-it is a 4.3 billion parameter instruction-tuned multimodal language model developed by Google, based on the Gemma 3 architecture. It is specifically trained for performance on medical text and image comprehension, utilizing a SigLIP image encoder pre-trained on diverse de-identified medical data. This model excels at medical applications involving text generation, visual question answering, and report generation from medical images like X-rays and histopathology slides.