Gemma 3 12B IT at Q4_K_M needs about 9.1 GB, more than the 8 GB of the RTX 4060: part of the model runs from system RAM, which slows generation. At Q3_K_M it does not fit in VRAM.
Q4_K_M · 8K context
The model needs about 9.1 GB but only 7.7 GB of VRAM is usable: 1.4 GB must be loaded into system RAM. It runs, 85% on the GPU, but more slowly.
~9.1 GB (8.5 GB–9.6 GB) needed · estimate
1.4 GB must be offloaded to system RAM
EstimateData updated 24 days agoSource: Run LLM Locally engine, Hugging Face config.json + model card
What makes this estimate less reliable
| Gemma 3 12B IT Q8_08K~14.1 GB · 3–5.5 tok/s estimate | CPU/GPU offload |
|---|---|
| Gemma 3 12B IT Q6_K8K~11.4 GB · 4.5–9 tok/s estimate | CPU/GPU offload |
| Gemma 3 12B IT Q5_K_M8K~10.2 GB · 6.5–12 tok/s estimate | CPU/GPU offload |
| Gemma 3 12B IT Q4_K_M8K~9.1 GB · 9.5–18 tok/s estimate | CPU/GPU offload |
| Gemma 3 12B IT IQ4_XS8K~8.3 GB · 15–25 tok/s estimate | CPU/GPU offload |
| Gemma 3 12B IT Q3_K_M8K~7.8 GB · 21–36 tok/s estimate | CPU/GPU offload |
Prices and links follow the market you choose.
NVIDIA
CompatibleRuns Gemma 3 12B IT Q4_K_M entirely in memory
AMD
ExcellentRuns Gemma 3 12B IT Q4_K_M entirely in memory
Unified memory PC
ExcellentRuns Gemma 3 12B IT Q4_K_M entirely in memory
No current offer imported for this market: the link opens a marketplace search.
