Updated daily at 05:00 UTC
Vision-language
MedGemma
Gemma 3 variants trained for medical text and image comprehension, released under the Health AI Developer Foundations terms.
By Google DeepMind.
Versions
4 versions
| Model | Maker | Kind | Released | Context | Input $/1M | Output $/1M |
|---|---|---|---|---|---|---|
| MedGemma 1.5 4BMedGemma · Open weights | Google DeepMind | Vision-language | 13 Jan 2026 | — | — | — |
| MedGemma 27B (multimodal)MedGemma · Open weights | Google DeepMind | Vision-language | 9 Jul 2025 | 131.1K | — | — |
| MedGemma 27B (text)MedGemma · Open weights | Google DeepMind | Vision-language | 20 May 2025 | 131.1K | — | — |
| MedGemma 4BMedGemma · Legacy · Open weights | Google DeepMind | Vision-language | 20 May 2025 | 131.1K | — | — |
