With 80 GB of VRAM, the H100 PCIe runs Phi-4 Mini Instruct 3.8B at Q4_K_M entirely on the GPU with an 8K context (about 4.5 GB needed). At Q8_0 it fits entirely.
Q4_K_M · 8K context
The model needs about 4.5 GB and fits in the 77.6 GB available, with 73.1 GB to spare.
~4.5 GB (4.1 GB–4.9 GB) needed · estimate
Fits entirely in VRAM · 73.1 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 Mini Instruct 3.8B Q8_08K~6 GB · 204–336 tok/s estimate | Excellent |
|---|---|
| Phi-4 Mini Instruct 3.8B Q6_K8K~5.1 GB · 232–399 tok/s estimate | Excellent |
| Phi-4 Mini Instruct 3.8B Q5_K_M8K~4.8 GB · 243–425 tok/s estimate | Excellent |
| Phi-4 Mini Instruct 3.8B Q4_K_M8K~4.5 GB · 258–460 tok/s estimate | Excellent |
| Phi-4 Mini Instruct 3.8B IQ4_XS8K~4.3 GB · 270–491 tok/s estimate | Excellent |
| Phi-4 Mini Instruct 3.8B Q3_K_M8K~4.2 GB · 275–504 tok/s estimate | Excellent |
Prices and links follow the market you choose.
NVIDIA
ExcellentRuns Phi-4 Mini Instruct 3.8B Q4_K_M entirely in memory
AMD
ExcellentRuns Phi-4 Mini Instruct 3.8B Q4_K_M entirely in memory
Unified memory PC
ExcellentRuns Phi-4 Mini Instruct 3.8B Q4_K_M entirely in memory
No current offer imported for this market: the link opens a marketplace search.
