Medgemma 27B Text IT — Hardware Requirements & GPU Compatibility
Chat
Google MedGemma 27B Text IT is a 27-billion parameter instruction-tuned model specialized for the medical domain, built on the Gemma architecture by Google. It is fine-tuned on medical and clinical text data to provide improved performance on healthcare-related tasks such as medical question answering, clinical reasoning, and health information summarization. The model requires a GPU with at least 24GB of VRAM for quantized inference. Its domain specialization makes it notably more capable than general models on clinical benchmarks, though it should not be used as a substitute for professional medical advice. Released under the Gemma license.
28.3K downloads
474 likesSpecifications
- Publisher
- Family
- Gemma
- Parameters
- 27.0B
- Release Date
- 2025-05-19
- License
- Other
Get Started
HuggingFace
How Much VRAM Does Medgemma 27B Text IT Need?
Select a quantization to see compatible GPUs below.
| Quantization | Bits | VRAM | + Context | File Size | Quality |
|---|---|---|---|---|---|
| BF16est. | 16.00 | 59.4 GB | — | 54.02 GB | Brain floating point 16 — preferred for training |
est.= calculated VRAM estimate; no published GGUF file found for that quantization yet. Other rows are verified against real community uploads.
Which GPUs Can Run Medgemma 27B Text IT?
BF16 · 59.4 GBMedgemma 27B Text IT (BF16) requires 59.4 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 78+ GB is recommended. No single GPU has enough memory — multi-GPU or cluster setups are needed.
Which Devices Can Run Medgemma 27B Text IT?
BF16 · 59.4 GB22 devices with unified memory can run Medgemma 27B Text IT, including NVIDIA DGX H100, NVIDIA DGX A100 640GB, Mac Studio (M3 Ultra, 96GB).
Runs great
— Plenty of headroom
Where to Download Medgemma 27B Text IT
Community quantizations of this model — GGUF for llama.cpp, Ollama, and LM Studio, plus AWQ/MLX variants where available.
Related Models
Frequently Asked Questions
- How much VRAM does Medgemma 27B Text IT need?
Medgemma 27B Text IT requires 59.4 GB of VRAM at BF16.
VRAM = Weights + KV Cache + Overhead
Weights = 27.0B × 16 bits ÷ 8 = 54 GB
KV Cache + Overhead ≈ 5.4 GB (at 2K context + ~0.3 GB framework)
VRAM usage by quantization
BF1659.4 GB- Can NVIDIA GeForce RTX 5090 run Medgemma 27B Text IT?
No — Medgemma 27B Text IT requires at least 59.4 GB at BF16, which exceeds the NVIDIA GeForce RTX 5090's 32 GB of VRAM.
- Can I run Medgemma 27B Text IT on a Mac?
Medgemma 27B Text IT requires at least 59.4 GB at BF16, which exceeds the unified memory of most consumer Macs. You would need a Mac Studio or Mac Pro with a high-memory configuration.
- Can I run Medgemma 27B Text IT locally?
Yes — Medgemma 27B Text IT can run locally on consumer hardware. At BF16 quantization it needs 59.4 GB of VRAM. Popular tools include Ollama, LM Studio, and llama.cpp.
- How fast is Medgemma 27B Text IT?
At BF16, Medgemma 27B Text IT can reach ~81 tok/s on AMD Instinct MI350X. Speed depends mainly on GPU memory bandwidth. Real-world results typically within ±20%.
tok/s = (bandwidth GB/s ÷ model GB) × efficiency
Example: NVIDIA B200 → 8000 ÷ 59.4 × 0.65 = ~88 tok/s
Estimated speed at BF16 (59.4 GB)
~88 tok/s~88 tok/s~81 tok/sReal-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
- What's the download size of Medgemma 27B Text IT?
At BF16, the download is about 54.02 GB.
- Which GPUs can run Medgemma 27B Text IT?
No single consumer GPU has enough VRAM to run Medgemma 27B Text IT at BF16 (59.4 GB). Multi-GPU or professional hardware is required.
- Which devices can run Medgemma 27B Text IT?
23 devices with unified memory can run Medgemma 27B Text IT at BF16 (59.4 GB), including ASUS Ascent GX10, Asus ROG Flow Z13 (2025, Ryzen AI Max+ 395, 128 GB), Beelink GTR9 Pro (Ryzen AI Max+ 395, 128 GB), Framework Desktop (Ryzen AI Max+ 395, 128 GB). Apple Silicon Macs use unified memory shared between CPU and GPU, making them well-suited for local LLM inference.