Google
Gemma 3 12B (12B parameters) requires approximately 14.0 GB of VRAM with Q4_K_M quantization. For the best balance of quality and speed, we recommend hardware with at least 17 GB of VRAM.
Get started
— copy & paste to run locallyCopy-paste commands to run Gemma 3 12B on your machine.
Run
ollama run gemma3:12bQuick specs
About this model
Related models
Quick picks
Best hardware
Run this model
Quantization options
No hardware detected — fit column shows raw VRAM estimates
| Quant | Bits | VRAM | Quality | Fit |
|---|---|---|---|---|
Q2_K | 2 | 4.7 GB | Low | — |
Q3_K_S | 3 | 5.9 GB | Low | — |
NVFP4 | 4 | 6.7 GB | Medium | — |
Q4_K_M | 4 | 7.3 GB | Medium | — |
Q5_K_M | 5 | 8.6 GB | High | — |
Q6_K | 6 | 9.8 GB | High | — |
Q8_0 | 8 | 12.8 GB | Very High | — |
F16 | 16 | 24.6 GB | Maximum | — |
Quality benchmarks
Coding
Reasoning
General
Source: official · 2025-03-12
Hardware compatibility
Computing compatibility...
Memory breakdown
Frequently asked questions
Gemma 3 12B (12B parameters) requires approximately 14.0 GB of VRAM with Q4_K_M quantization. Lower quantizations like Q4_K_M use less memory but may reduce quality.
Yes, RX 7600 XT 16GB can run Gemma 3 12B with a compatibility score of 78/100. It provides 16 GB of memory and achieves approximately 18.2 tokens per second.
The recommended quantization for Gemma 3 12B is Q4_K_M, which offers the best balance between model quality and memory efficiency. Higher quantizations preserve more quality but require more VRAM.
The top recommended hardware for Gemma 3 12B: RTX A4500 20GB (score: 86/100), RTX 3090 24GB (score: 86/100), RTX 3090 Ti 24GB (score: 86/100). These provide the best combination of memory, bandwidth, and compute for running this model locally.
Yes, Gemma 3 12B is well-suited for chat as well as coding. It was designed with these use cases in mind.
See also
