Qwen2.5-Coder 3B
Qwen/Qwen2.5-Coder-3B-Instruct
Parameters
3.1B
Context
128K
Downloads
111.8K
Likes
123
Architecture
transformer
GPU Requirements
VRAM / RAM / disk estimates by quantization and usage scenario
| Quantization | Minimum | Recommended | Production |
|---|---|---|---|
| FP1616bit | 12.3 GB RAM 32.0 · Disk 48.0 | 28.3 GB RAM 42.5 · Disk 74.0 | 125.6 GB RAM 188.4 · Disk 113.0 |
| FP88bit | 8.6 GB RAM 32.0 · Disk 43.5 | 24.5 GB RAM 36.8 · Disk 69.5 | 121.6 GB RAM 182.4 · Disk 108.5 |
| INT88bit | 8.6 GB RAM 32.0 · Disk 43.5 | 24.5 GB RAM 36.8 · Disk 69.5 | 121.6 GB RAM 182.4 · Disk 108.5 |
| AWQ4bit | 6.8 GB RAM 32.0 · Disk 41.3 | 22.7 GB RAM 34.0 · Disk 67.3 | 119.7 GB RAM 179.5 · Disk 106.3 |
| GPTQ4bit | 6.9 GB RAM 32.0 · Disk 41.4 | 22.7 GB RAM 34.1 · Disk 67.4 | 119.7 GB RAM 179.5 · Disk 106.4 |
| GGUF4bit | 6.9 GB RAM 32.0 · Disk 41.4 | 22.8 GB RAM 34.1 · Disk 67.4 | 119.7 GB RAM 179.6 · Disk 106.4 |
GPU Recommendations
1× RTX 5090
32 GB VRAM
≈ 1194.7 tok/s
Effective production throughput ≈ 836.3 tok/s
score 0.6857
1× RTX 3090
24 GB VRAM
≈ 624 tok/s
Effective production throughput ≈ 436.8 tok/s
score 0.6428
8× H200 SXM
1128 GB VRAM
≈ 10240 tok/s
Effective production throughput ≈ 7168 tok/s
score 0.2197
1× RTX 4090
24 GB VRAM
≈ 672 tok/s
Effective production throughput ≈ 470.4 tok/s
score 0.6412
💬 Community — real deployment experience
0 Articles · 0 Benchmarks
GPUs that fit (single card)
RTX 3090
24 GB VRAM
$88.4
/mo · vast
A10
24 GB VRAM
$171
/mo · vast
L4
24 GB VRAM
$185
/mo · vast
RTX 4090
24 GB VRAM
$190
/mo · tensordock
RTX A6000
48 GB VRAM
$292
/mo · vast