With 24 GB of VRAM, the Tesla P40 runs Phi-4 14B at Q4_K_M entirely on the GPU with an 8K context (about 11.2 GB needed). At Q8_0 it fits entirely.
Q4_K_M · 8K context
The model needs about 11.2 GB and fits in the 23.3 GB available, with 12 GB to spare.
~11.2 GB (10.8 GB–11.8 GB) needed · estimate
Fits entirely in VRAM · 12 GB left
EstimateData updated 24 days agoSource: Run LLM Locally engine, Hugging Face config.json + model card
What makes this estimate less reliable
| Phi-4 14B Q8_08K~17.6 GB · 16–20 tok/s estimate | Excellent |
|---|---|
| Phi-4 14B Q6_K8K~14.2 GB · 20–26 tok/s estimate | Excellent |
| Phi-4 14B Q5_K_M8K~12.8 GB · 22–29 tok/s estimate | Excellent |
| Phi-4 14B Q4_K_M8K~11.2 GB · 25–34 tok/s estimate | Excellent |
| Phi-4 14B IQ4_XS8K~10.2 GB · 28–38 tok/s estimate | Excellent |
| Phi-4 14B Q3_K_M8K~9.6 GB · 30–40 tok/s estimate | Excellent |
Prices and links follow the market you choose.
NVIDIA
ExcellentRuns Phi-4 14B Q4_K_M entirely in memory
AMD
ExcellentRuns Phi-4 14B Q4_K_M entirely in memory
Apple Silicon
CompatibleRuns Phi-4 14B Q4_K_M entirely in memory
Unified memory PC
ExcellentRuns Phi-4 14B Q4_K_M entirely in memory
No current offer imported for this market: the link opens a marketplace search.
