What GPU do I need to run google/medgemma-4b-it?
4.3B parameters, published in BF16. View on Hugging FaceGated
Set up medgemma-4b-itmedgemma-4b-it is published by google on Hugging Face, with 1,003,531 downloads and 1,041 likes to date. It's a Gemma3ForConditionalGeneration model built for image-text-to-text, published natively in BF16, and gated: you'll need to accept the model's terms on Hugging Face before downloading weights.
VRAM required & cheapest live GPU fit
Required VRAM = weight size at each precision times a flat 1.2 overhead for activations and fragmentation. The KV-cache grows with context and is not in that factor; it is listed per model below where the architecture is published. Full formula and assumptions: methodology.
| Precision | Weight size | Required VRAM | Cheapest live fit | GPUs needed | Est. $/hr (full fit) |
|---|---|---|---|---|---|
| BF16 | 8.0 GB | 9.6 GB | RTX A5000 | 1 | $0.176/hr |
| FP8 (quantized) | 4.0 GB | 4.8 GB | RTX 4000 SFF Ada | 1 | $0.198/hr |
| INT4 (quantized) | 2.0 GB | 2.4 GB | RTX 3070 | 1 | $0.143/hr |
A GPU is only matched to a row if its hardware supports that precision, and the primary recommendation is always a single-GPU fit when one exists.
INT4 requires a quantized checkpoint actually published for this model, check its Hugging Face page before relying on this row.
Cheapest way to run medgemma-4b-it at its published (BF16) precision: 1× RTX A5000, at $0.176/hr per GPU ($0.176/hr total). Quantizing to FP8 or INT4 (rows above) can cost less, but requires a compatible quantized checkpoint to exist for this model.
medgemma-4b-it: KV cache by context length
The KV cache is the memory the attention layers hold for every token of context, on top of the weights. It is not part of the flat 1.2x overhead in the table above, grows with context length and with every concurrent request, and is why a long-context deployment needs more VRAM than the table shows.
Sliding-window attention: windowed layers stop growing at the window, the full-attention layers keep growing. Cached per token: 139,264 bytes at 16-bit while the context is inside the window.
| Context | KV cache, one sequence |
|---|---|
| 4K | 0.19 GB |
| 32K | 0.74 GB |
| 128K | 2.6 GB |
Computed from the layer, head and window counts in the model's own configuration (unsloth/medgemma-4b-it), at 16 bits per cached value. Each concurrent sequence needs its own cache, on top of the weights.
medgemma-4b-it: common questions
Does medgemma-4b-it fit on a 12 GB GPU?
Yes. At BF16 it needs 9.6 GB of VRAM, so a 12 GB card holds it with 2.4 GB to spare. An 8 GB card is not enough for it at BF16.
Do I need approval to download medgemma-4b-it?
Yes. google gates this repository on Hugging Face, so you have to accept its terms with a signed-in account before the weights will download. It does not change the 9.6 GB the model needs once you have them.
How many copies of medgemma-4b-it fit on one RTX A5000?
2, by VRAM alone. That card carries 24.0 GB and one copy needs 9.6 GB at BF16, on a live rate of $0.176/hr for the whole card. Throughput is not modelled here, so 2 copies is not 2 times the requests served.
What is the least VRAM medgemma-4b-it can run in?
2.4 GB, at INT4 (quantized), which fits a 6 GB card, against 9.6 GB at BF16. That row assumes an INT4 checkpoint has actually been published for this model, so check its Hugging Face page before planning around it.
Weight-to-VRAM math, the fit rules, and how live prices are normalized: full methodology.
More MedGemma models
- medgemma-1.5-4b-it (4.3B, BF16)
- medgemma-27b-it (28.8B, BF16)
- medgemma-27b-text-it (27.0B, BF16)
Alternatives at this size
Other models for image-text-to-text within about a third of medgemma-4b-it's 4.3B parameters, from other model lines.
- Qwen3.5-4B (4.7B, BF16)
- Unlimited-OCR (3.3B, BF16)
- chandra-ocr-2 (5.3B, BF16)
- DeepSeek-OCR (3.3B, BF16)
- Phi-3.5-vision-instruct (4.1B, BF16)
More on medgemma-4b-it
Fits on a 12 GB GPU at BF16: every model that fits in 12 GB.
Fits on an 8 GB GPU at FP8: every model that fits in 8 GB.
Best vision-language models: how medgemma-4b-it ranks against the rest.
Related reading: RTX A5000 pricing and specs, How much VRAM you need for LLMs, Serving LLMs with vLLM, vLLM vs TensorRT-LLM vs SGLang, and Best GPU for LLM inference.
