Home / Models / Phi-4 Mini Instruct
Phi-4 Mini Instruct
3.8B parameters · native context 131,072 · phi3
Estimated VRAM requirement
Weight memory uses parameter count × bits per weight, plus a quantization overhead factor. KV cache uses standard GQA formulas. Values are calculated estimates, not measured allocations.
| Quantization | Estimated weights | 2K total | 4K total | 8K total | 16K total | 32K total | 64K total | 128K total |
|---|---|---|---|---|---|---|---|---|
| FP16 | 7.3 GB | 8.9 GB | 9.1 GB | 9.6 GB | 10.6 GB | 12.6 GB | 16.6 GB | 24.6 GB |
| Q8 | 4.1 GB | 5.4 GB | 5.7 GB | 6.2 GB | 7.2 GB | 9.2 GB | 13.2 GB | 21.2 GB |
| Q6 | 3.2 GB | 4.5 GB | 4.7 GB | 5.2 GB | 6.2 GB | 8.2 GB | 12.2 GB | 20.2 GB |
| Q5 | 2.7 GB | 3.9 GB | 4.2 GB | 4.7 GB | 5.7 GB | 7.7 GB | 11.7 GB | 19.7 GB |
| Q4 | 2.3 GB | 3.4 GB | 3.7 GB | 4.2 GB | 5.2 GB | 7.2 GB | 11.2 GB | 19.2 GB |
| Q3 | 1.8 GB | 2.9 GB | 3.2 GB | 3.7 GB | 4.7 GB | 6.7 GB | 10.7 GB | 18.7 GB |
Find a GPU for this model
Run at the quantization and context selected below. Defaults are Q4 and 8K when you do not change the controls. Results are full-VRAM fits only — not a speed ranking.
- Run at
- Q4 · 8K context
- Required VRAM
- 4.2 GB
No fresh matched Amazon price is currently available for a full-VRAM fit. See compatible GPUs · Workstations · Rent a server · Buy vs rent.
Lowest-cost full-VRAM option
Sorted by lowest current Amazon price among GPUs that fit entirely in VRAM. This is not a performance ranking.
No compatible GPU currently has a fresh Amazon price.
More VRAM headroom
Sorted by leftover usable VRAM after the calculated requirement. Headroom is not tokens/sec.
NVIDIA B200 192 GB
- GPU VRAM
- 192 GB advertised
- Required VRAM
- 4.2 GB
- Usable GPU VRAM
- 172.8 GB
- VRAM headroom
- 168.6 GB — leftover usable memory after the calculated requirement, not a speed ranking
- Compatible
- Yes — fits entirely in VRAM
- Product SKUs
- 0 mapped · 0 Amazon-matched
- Amazon
- Check availability
NVIDIA H200 141 GB
- GPU VRAM
- 141 GB advertised
- Required VRAM
- 4.2 GB
- Usable GPU VRAM
- 126.9 GB
- VRAM headroom
- 122.7 GB — leftover usable memory after the calculated requirement, not a speed ranking
- Compatible
- Yes — fits entirely in VRAM
- Product SKUs
- 0 mapped · 0 Amazon-matched
- Amazon
- Check availability
NVIDIA GH200 96 GB
- GPU VRAM
- 96 GB advertised
- Required VRAM
- 4.2 GB
- Usable GPU VRAM
- 86.4 GB
- VRAM headroom
- 82.2 GB — leftover usable memory after the calculated requirement, not a speed ranking
- Compatible
- Yes — fits entirely in VRAM
- Product SKUs
- 0 mapped · 0 Amazon-matched
- Amazon
- Check availability
Other compatible GPUs
AMD Radeon RX 470 8 GB
- GPU VRAM
- 8 GB advertised
- Required VRAM
- 4.2 GB
- Usable GPU VRAM
- 7.2 GB
- VRAM headroom
- 3.0 GB — leftover usable memory after the calculated requirement, not a speed ranking
- Compatible
- Yes — fits entirely in VRAM
- Product SKUs
- 1 mapped · 0 Amazon-matched
- Amazon
- Check availability
NVIDIA Quadro P4000 8 GB
- GPU VRAM
- 8 GB advertised
- Required VRAM
- 4.2 GB
- Usable GPU VRAM
- 7.2 GB
- VRAM headroom
- 3.0 GB — leftover usable memory after the calculated requirement, not a speed ranking
- Compatible
- Yes — fits entirely in VRAM
- Product SKUs
- 6 mapped · 2 Amazon-matched
- Amazon
- Check price on Amazon
AMD Radeon RX 580 8 GB
- GPU VRAM
- 8 GB advertised
- Required VRAM
- 4.2 GB
- Usable GPU VRAM
- 7.2 GB
- VRAM headroom
- 3.0 GB — leftover usable memory after the calculated requirement, not a speed ranking
- Compatible
- Yes — fits entirely in VRAM
- Product SKUs
- 13 mapped · 4 Amazon-matched
- Amazon
- Check price on Amazon
Compatible GPUs
Status is a calculated memory fit against usable VRAM (advertised × 0.9). Open a GPU to see product SKUs and Amazon listings.
Compare GPUs for this model
Published chip comparisons where both GPUs fully fit this model at Q4 / 8K. Opens with this model preselected; that query is not indexed.
How fast can this model run?
llama.cpp CUDA decode at Q4 under llama-bench tg128 conditions. Not a ranking. Open a pair for prefill and methodology.
| GPU | Decode | State | Confidence |
|---|---|---|---|
| NVIDIA RTX 5090 32 GB | ~475 tok/s | Calibrated estimate | MEDIUM |
| NVIDIA RTX 4090 24 GB | ~299 tok/s | Calibrated estimate | MEDIUM |
| NVIDIA RTX 5080 16 GB | ~292 tok/s | Calibrated estimate | MEDIUM |
| NVIDIA RTX 5070 Ti 16 GB | ~288 tok/s | Calibrated estimate | MEDIUM |
| NVIDIA RTX 3090 Ti 24 GB | ~272 tok/s | Calibrated estimate | MEDIUM |
| NVIDIA RTX 3090 24 GB | ~256 tok/s | Calibrated estimate | MEDIUM |
| NVIDIA RTX 4080 SUPER 16 GB | ~233 tok/s | Calibrated estimate | MEDIUM |
| NVIDIA RTX A6000 48 GB | ~229 tok/s | Calibrated estimate | MEDIUM |
| NVIDIA RTX 4080 16 GB | ~227 tok/s | Calibrated estimate | MEDIUM |
| NVIDIA RTX 3080 10 GB | ~221 tok/s | Calibrated estimate | MEDIUM |
Sources
Model specifications: official repository microsoft/Phi-4-mini-instruct · revision main. Compatibility: RigForAI calculator v1.0.0 (estimated VRAM). GPU products: Icecat. Amazon is a separate affiliate match and is not used as a spec source.