Qwen3 14B at Q4_K_M needs about 11 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 11 GB but only 7.7 GB of VRAM is usable: 3.3 GB must be loaded into system RAM. It runs, 70% on the GPU, but more slowly.
~11 GB (10.5 GB–11.5 GB) needed · estimate
3.3 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
| Qwen3 14B Q8_08K~17.5 GB · 2–3.5 tok/s estimate | Possible but slow |
|---|---|
| Qwen3 14B Q6_K8K~14 GB · 3–5.5 tok/s estimate | CPU/GPU offload |
| Qwen3 14B Q5_K_M8K~12.5 GB · 4–7 tok/s estimate | CPU/GPU offload |
| Qwen3 14B Q4_K_M8K~11 GB · 5–10 tok/s estimate | CPU/GPU offload |
| Qwen3 14B IQ4_XS8K~10.1 GB · 6.5–13 tok/s estimate | CPU/GPU offload |
| Qwen3 14B Q3_K_M8K~9.3 GB · 8.5–16 tok/s estimate | CPU/GPU offload |
Prices and links follow the market you choose.
NVIDIA
CompatibleRuns Qwen3 14B Q4_K_M entirely in memory
AMD
ExcellentRuns Qwen3 14B Q4_K_M entirely in memory
Apple Silicon
CompatibleRuns Qwen3 14B Q4_K_M entirely in memory
Unified memory PC
ExcellentRuns Qwen3 14B Q4_K_M entirely in memory
No current offer imported for this market: the link opens a marketplace search.
