Run MaziyarPanahi/gemma-3-12b-it-GGUF locally
MaziyarPanahi/gemma-3-12b-it-GGUF is a large instruction-tuned chat model with 11.77 billion parameters, built on the gemma3 architecture. It has been downloaded 151,517 times.
To run MaziyarPanahi/gemma-3-12b-it-GGUF locally at a 4,096-token context, its quantized versions need between 6.65 GB (Q2_K, lowest quality) and 24.13 GB (GGUF, highest quality) of memory, weights plus KV cache and a system margin included.
For most users the best balance is Q8_0, needing about 13.86 GB. That means MaziyarPanahi/gemma-3-12b-it-GGUF fits entirely in the VRAM of an 8 GB GPU or larger, running fully on the GPU.
All quantizations
| Quant. | Bits | Quality | Weights | KV | Total | Speed~ | Verdict |
|---|---|---|---|---|---|---|---|
| Q2_K | 3.24 | Low | 4.44 GB | 1.41 GB | 6.65 GB | 11.3 t/s | Offload |
| Q3_K_S | 3.71 | Fair | 5.08 GB | 1.41 GB | 7.29 GB | 9.8 t/s | Offload |
| Q3_K_M | 4.09 | Fair | 5.6 GB | 1.41 GB | 7.8 GB | 8.9 t/s | Offload |
| Q3_K_L | 4.41 | Good | 6.03 GB | 1.41 GB | 8.24 GB | 8.3 t/s | Offload |
| Q4_K_S | 4.72 | Good | 6.46 GB | 1.41 GB | 8.67 GB | 7.7 t/s | Offload |
| Q4_K_M | 4.96 | Good | 6.8 GB | 1.41 GB | 9.01 GB | 7.4 t/s | Offload |
| Q5_K_S | 5.6 | Very good | 7.67 GB | 1.41 GB | 9.87 GB | 6.5 t/s | Offload |
| Q5_K_M | 5.74 | Very good | 7.86 GB | 1.41 GB | 10.07 GB | 6.4 t/s | Offload |
| Q6_K | 6.57 | Excellent | 9.0 GB | 1.41 GB | 11.2 GB | 5.6 t/s | Offload |
| Q8_0 | 8.51 | Excellent | 11.65 GB | 1.41 GB | 13.86 GB | 4.3 t/s | Offload |
| GGUF | 16.01 | Excellent | 21.92 GB | 1.41 GB | 24.13 GB | — | Insufficient |
KV cache computed from the model's exact architecture. Speed is a rough estimate bounded by memory bandwidth.
Frequently asked questions
How much VRAM do you need to run MaziyarPanahi/gemma-3-12b-it-GGUF?
You need about 7.8 GB of VRAM to run MaziyarPanahi/gemma-3-12b-it-GGUF entirely on the GPU using the Q3_K_M quantization (at a 4,096-token context). Smaller quantizations lower the requirement at the cost of quality.
Can I run MaziyarPanahi/gemma-3-12b-it-GGUF on an 8 GB GPU?
Yes. With 8 GB of VRAM you can run MaziyarPanahi/gemma-3-12b-it-GGUF fully on the GPU using Q3_K_M (about 7.8 GB).
Can I run MaziyarPanahi/gemma-3-12b-it-GGUF on a 16 GB GPU?
Yes. With 16 GB of VRAM you can run MaziyarPanahi/gemma-3-12b-it-GGUF fully on the GPU using Q8_0 (about 13.86 GB).
Can I run MaziyarPanahi/gemma-3-12b-it-GGUF on a 24 GB GPU?
Yes. With 24 GB of VRAM you can run MaziyarPanahi/gemma-3-12b-it-GGUF fully on the GPU using Q8_0 (about 13.86 GB).
What is the best quantization for MaziyarPanahi/gemma-3-12b-it-GGUF?
If memory allows, higher bits-per-weight means better quality. A common sweet spot is a Q4_K_M or Q5_K_M quantization, which keeps most of the quality while roughly halving the memory versus 8-bit. Pick the highest quantization that still fits in your VRAM.