Phi-3 Mini 3.8B
microsoft/Phi-3-mini-128k-instruct
Parameters
3.8B
Context
128K
Downloads
228.4K
Likes
1.7K
Architecture
transformer
GPU Requirements
VRAM / RAM / disk estimates by quantization and usage scenario
| Quantization | Minimum | Recommended | Production |
|---|---|---|---|
| FP1616bit | 13.9 GB RAM 32.0 · Disk 50.0 | 30.1 GB RAM 45.1 · Disk 76.0 | 127.4 GB RAM 191.1 · Disk 115.0 |
| FP88bit | 9.4 GB RAM 32.0 · Disk 44.5 | 25.4 GB RAM 38.1 · Disk 70.5 | 122.5 GB RAM 183.7 · Disk 109.5 |
| INT88bit | 9.4 GB RAM 32.0 · Disk 44.5 | 25.4 GB RAM 38.1 · Disk 70.5 | 122.5 GB RAM 183.7 · Disk 109.5 |
| AWQ4bit | 7.3 GB RAM 32.0 · Disk 41.9 | 23.1 GB RAM 34.7 · Disk 67.9 | 120.1 GB RAM 180.2 · Disk 106.9 |
| GPTQ4bit | 7.3 GB RAM 32.0 · Disk 41.9 | 23.2 GB RAM 34.7 · Disk 67.9 | 120.2 GB RAM 180.3 · Disk 106.9 |
| GGUF4bit | 7.4 GB RAM 32.0 · Disk 42.0 | 23.2 GB RAM 34.8 · Disk 68.0 | 120.2 GB RAM 180.3 · Disk 107.0 |
GPU Recommendations
1× RTX 5090
32 GB VRAM
≈ 973.9 tok/s
Effective production throughput ≈ 681.7 tok/s
score 0.6902
1× RTX 3090
24 GB VRAM
≈ 508.7 tok/s
Effective production throughput ≈ 356.1 tok/s
score 0.6418
8× H200 SXM
1128 GB VRAM
≈ 8347.8 tok/s
Effective production throughput ≈ 5843.5 tok/s
score 0.2197
1× RTX 4090
24 GB VRAM
≈ 547.8 tok/s
Effective production throughput ≈ 383.5 tok/s
score 0.6418
💬 Community — real deployment experience
0 Articles · 0 Benchmarks
GPUs that fit (single card)
RTX 3090
24 GB VRAM
$76.5
/mo · vast
RTX 4090
24 GB VRAM
$99.0
/mo · vast
RTX 4090D
24 GB VRAM
$117
/mo · vast
L4
24 GB VRAM
$147
/mo · vast
A10
24 GB VRAM
$176
/mo · vast