With 24 GB of VRAM, the Tesla P40 runs Qwen3 14B at Q4_K_M entirely on the GPU with an 8K context (about 11 GB needed). At Q8_0 it fits entirely.
Q4_K_M · 8K context
The model needs about 11 GB and fits in the 23.3 GB available, with 12.3 GB to spare.
~11 GB (10.5 GB–11.5 GB) needed · estimate
Fits entirely in VRAM · 12.3 GB left
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 · 16–21 tok/s estimate | Excellent |
|---|---|
| Qwen3 14B Q6_K8K~14 GB · 20–26 tok/s estimate | Excellent |
| Qwen3 14B Q5_K_M8K~12.5 GB · 23–30 tok/s estimate | Excellent |
| Qwen3 14B Q4_K_M8K~11 GB · 26–34 tok/s estimate | Excellent |
| Qwen3 14B IQ4_XS8K~10.1 GB · 28–38 tok/s estimate | Excellent |
| Qwen3 14B Q3_K_M8K~9.3 GB · 31–41 tok/s estimate | Excellent |
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.
