Medgemma 27b It needs about 21.2 GB of memory at Q4_K_M (a 16.5 GB file) with an 8K context window, so it fits on a 24 GB GPU. This repository has 26 quants, from 6.3 GB (IQ1_S) to 54.0 GB (F16). Longer context needs more: at 32K tokens the KV cache alone adds about 16.6 GB.
Medgemma 27b It – 27B multimodal model for medical text and image tasks
Medgemma 27b It is a 27B parameter Gemma3 variant trained for medical text and image comprehension. It supports embedding and vision abilities. The model is available in multiple quantizations for different hardware budgets.
Good at
- Medical text generation
- Medical image captioning
- Embedding extraction
Limits
- No vision support on low‑memory quantizations
- Large VRAM required for full precision
- License governed by Health AI Developer Foundations terms
Best for
- Healthcare app prototyping
- Medical research data analysis
- Clinical decision support tools
Who it’s for. Developers building healthcare AI solutions who need a large multimodal model with medical domain specialization. The model is suitable for research and production environments that can meet its memory requirements.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| IQ1_S | 6.3 GB | 10.9 GB | Too big | Get file → | |
| IQ1_M | 6.8 GB | 11.5 GB | Too big | Get file → | |
| IQ2_XXS | 7.7 GB | 12.3 GB | Too big | Get file → | |
| IQ2_XS | 8.4 GB | 13.1 GB | Too big | Get file → | |
| IQ2_S | 8.8 GB | 13.4 GB | Too big | Get file → | |
| IQ2_M | 9.5 GB | 14.2 GB | Too big | Get file → | |
| Q2_K_S | 9.8 GB | 14.4 GB | Too big | Get file → | |
| Q2_K | 10.5 GB | 15.2 GB | Too big | Get file → | |
| IQ3_XXS | 10.7 GB | 15.4 GB | Too big | Get file → | |
| IQ3_XS | 11.6 GB | 16.2 GB | Too big | Get file → | |
| IQ3_S | 12.2 GB | 16.8 GB | Too big | Get file → | |
| Q3_K_S | 12.2 GB | 16.8 GB | Too big | Get file → | |
| IQ3_M | 12.5 GB | 17.2 GB | Too big | Get file → | |
| Q3_K_M | 13.4 GB | 18.1 GB | Too big | Get file → | |
| Q3_K_L | 14.5 GB | 19.2 GB | Too big | Get file → | |
| IQ4_XS | 14.8 GB | 19.4 GB | Too big | Get file → | |
| IQ4_NL | 15.6 GB | 20.2 GB | Too big | Get file → | |
| Q4_0 | 15.6 GB | 20.3 GB | Too big | Get file → | |
| Q4_K_S | 15.7 GB | 20.3 GB | Too big | Get file → | |
| Q4_K_M | 16.5 GB | 21.2 GB | Too big | Get file → | |
| Q5_K_S | 18.8 GB | 23.4 GB | Too big | Get file → | |
| Q5_0 | 18.8 GB | 23.5 GB | Too big | Get file → | |
| Q5_K_M | 19.3 GB | 23.9 GB | Too big | Get file → | |
| Q6_K | 22.2 GB | 26.8 GB | Too big | Get file → | |
| Q8_0 | 28.7 GB | 33.4 GB | Too big | Get file → | |
| F16 | 54.0 GB | 58.7 GB | Too big | Get file → |
For 8 GB VRAM use IQ1_S or IQ1_M, which estimate 10.9 GB or 11.5 GB memory at 8 k context. For 16 GB VRAM use Q4_K_M or Q5_K_M, estimating 21.2 GB or 23.9 GB. Full precision (F16) requires 58.7 GB and is only feasible on high‑end GPUs.
What GPU do you need for Medgemma 27b It?
These figures add the file size, an fp16 KV cache at an 8K context window (worked out from this model’s own header) and about 0.5 GB for the runtime. They are estimates, so leave a little headroom, and expect to use less context or a smaller quant if you are right at the limit.
- 4 GB VRAM: no quant fits at an 8K context.
- 6 GB VRAM: no quant fits at an 8K context.
- 8 GB VRAM: no quant fits at an 8K context.
- 12 GB VRAM: the best quant that fits is IQ1_M (6.8 GB file, about 11.5 GB in use).
- 16 GB VRAM: the best quant that fits is IQ3_XXS (10.7 GB file, about 15.4 GB in use).
- 24 GB VRAM: the best quant that fits is Q5_K_M (19.3 GB file, about 23.9 GB in use).
- 32 GB VRAM: the best quant that fits is Q6_K (22.2 GB file, about 26.8 GB in use).
- 48 GB VRAM: the best quant that fits is Q8_0 (28.7 GB file, about 33.4 GB in use).
Which Medgemma 27b It quant should you choose?
Q4_K_M is the usual starting point for Medgemma 27b It: it balances file size and quality well. Here it is a 16.5 GB file.
The Q8_0 file is 28.7 GB, about 1.7 times the size of Q4_K_M. It keeps the weights closer to the original, which only matters if you have the memory to spare.
If memory is tight, the smallest file is IQ1_S at 6.3 GB, with the quality trade-offs that come with fewer bits. Use it only when memory is very tight. Expect a clear drop in quality, which is usually worse on small models and easier to live with on large ones.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about Medgemma 27b It
How much VRAM does Medgemma 27b It need?
About 21.2 GB at Q4_K_M with an 8K context window. That is the 16.5 GB file, plus the KV cache and roughly 0.5 GB of runtime overhead. Smaller quants need less and longer context needs more.
Can I run Medgemma 27b It on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 10.9 GB, so you would need a larger GPU or CPU offloading.
What is the file size of Medgemma 27b It GGUF?
It depends on the quant: IQ1_S is 6.3 GB, IQ1_M is 6.8 GB, IQ2_XXS is 7.7 GB, IQ2_XS is 8.4 GB, IQ2_S is 8.8 GB, IQ2_M is 9.5 GB and 20 more.
What context length does Medgemma 27b It support?
The GGUF header lists a trained context length of 131,072 tokens (128K). The KV cache grows with the context you use: about 4.2 GB at 8K and 16.6 GB at 32K.
Where can I download Medgemma 27b It GGUF?
The files are on Hugging Face in the gabriellarson/medgemma-27b-it-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.
What license does Medgemma 27b It use?
Hugging Face lists the license as other. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.
Can I run Medgemma 27b It on 8 GB VRAM?
Only the IQ1_S and IQ1_M quantizations are close to 8 GB VRAM, but they still estimate 10.9 GB and 11.5 GB memory at 8 k context. A larger GPU is recommended.
Does the model support vision tasks?
Yes, Medgemma 27b It includes vision capabilities via a SigLIP image encoder.
What license applies to this model?
Use is governed by the Health AI Developer Foundations terms of use.