Gemma 3 270M — Hardware Requirements & GPU Compatibility
Chat
Google Gemma 3 270M is a 270-million parameter base (pretrained) model from Google's Gemma 3 family. It is an experimental release intended for research, fine-tuning, and exploring the capabilities of ultra-small language models. The model runs on virtually any hardware with negligible resource requirements. Released under the Gemma license.
83.4K downloads
995 likesSpecifications
- Publisher
- Family
- Gemma
- Parameters
- 270M
- Release Date
- 2025-08-14
- License
- Gemma Terms
Get Started
HuggingFace
How Much VRAM Does Gemma 3 270M Need?
Select a quantization to see compatible GPUs below.
| Quantization | Bits | VRAM | + Context | File Size | Quality |
|---|---|---|---|---|---|
| Q8_0 | 8.00 | 0.3 GB | — | 0.27 GB | 8-bit quantization, near-lossless |
Which GPUs Can Run Gemma 3 270M?
Q8_0 · 0.3 GBGemma 3 270M (Q8_0) requires 0.3 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 1+ GB is recommended. 35 GPUs can run it, including NVIDIA GeForce RTX 5090, NVIDIA GeForce RTX 3090 Ti.
Runs great
— Plenty of headroomWhich Devices Can Run Gemma 3 270M?
Q8_0 · 0.3 GB33 devices with unified memory can run Gemma 3 270M, including NVIDIA DGX H100, NVIDIA DGX A100 640GB.
Runs great
— Plenty of headroomRelated Models
Derivatives (5)
Frequently Asked Questions
- How much VRAM does Gemma 3 270M need?
Gemma 3 270M requires 0.3 GB of VRAM at Q8_0.
VRAM = Weights + KV Cache + Overhead
Weights = 270M × 8 bits ÷ 8 = 0.3 GB
VRAM usage by quantization
Q8_00.3 GB- Can I run Gemma 3 270M on a Mac?
Gemma 3 270M requires at least 0.3 GB at Q8_0, 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 Gemma 3 270M locally?
Yes — Gemma 3 270M can run locally on consumer hardware. At Q8_0 quantization it needs 0.3 GB of VRAM. Popular tools include Ollama, LM Studio, and llama.cpp.
- How fast is Gemma 3 270M?
At Q8_0, Gemma 3 270M can reach ~9717 tok/s on AMD Instinct MI300X. On NVIDIA GeForce RTX 4090: ~2184 tok/s. Speed depends mainly on GPU memory bandwidth. Real-world results typically within ±20%.
tok/s = (bandwidth GB/s ÷ model GB) × efficiency
Example: AMD Instinct MI300X → 5300 ÷ 0.3 × 0.55 = ~9717 tok/s
Estimated speed at Q8_0 (0.3 GB)
AMD Instinct MI300X~9717 tok/sNVIDIA GeForce RTX 4090~2184 tok/sNVIDIA H100 SXM~7263 tok/sAMD Instinct MI250X~6008 tok/sReal-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
- What's the download size of Gemma 3 270M?
At Q8_0, the download is about 0.27 GB.