Gemma 3 270M
Gemma 3 270M is an open-weights language model from Google, released August 14, 2025. It has 0.3B parameters (dense).
Today, we're adding a new, highly specialized tool to the Gemma 3 toolkit: Gemma 3 270M, a compact, 270-million parameter model designed from the ground up for task-specific fine-tuning with strong instruction-following and text structuring capabilities already trained in.
Check it against your own PC →
Can you run Gemma 3 270M locally?
The smallest common GPU it runs on is 6 GB of VRAM (with 32 GB of RAM): Q6_K, fully on GPU, roughly 340 tokens per second. Estimates assume 32 GB of DDR5 RAM, a 16K context and llama.cpp-style GGUF quantization. Enter your exact hardware in the live app.
| GPU memory | Verdict | Best quant | VRAM | RAM | tok/s |
|---|---|---|---|---|---|
| 6 GB VRAM | Runs on GPU | Q6_K (≈ lossless) | 2.0 GB | – | ~340 |
| 8 GB VRAM | Runs on GPU | Q6_K (≈ lossless) | 2.0 GB | – | ~420 |
| 12 GB VRAM | Runs on GPU | Q6_K (≈ lossless) | 2.0 GB | – | ~490 |
| 16 GB VRAM | Runs on GPU | Q6_K (≈ lossless) | 2.0 GB | – | ~580 |
| 24 GB VRAM | Runs on GPU | Q6_K (≈ lossless) | 2.0 GB | – | ~720 |
| 32 GB VRAM | Runs on GPU | Q6_K (≈ lossless) | 2.0 GB | – | ~830 |
Ready-made quantizations on Hugging Face: Rgjur/gemma-3-270m-gguf · gguf-org/gemma-3-270m-gguf · pixasocial/survival-uncensored-gemma-270m · ggml-org/gemma-3-270m-GGUF · QLNI/embeddings-are-overpriced · QuantFactory/gemma-3-270m-it-GGUF
Benchmark results
Epoch AI has not published results for Gemma 3 270M yet. New models are usually evaluated within days to weeks.
Links
Announcement · Hugging Face: google/gemma-3-270m
More from Google
Gemini 4 Carbon · Gemini 4 Argon · Gemini 3.8 Flash · Gemini 3.7 Flash · Gemini 3.6 Flash · Gemini 3.5 Flash-Lite · Gemini 3.5 Flash Cyber · Gemma 4 E4B