Qwen/Qwen2.5-0.5B-Instruct
Qwen/Qwen2.5-0.5B-Instruct
Parameters
367.3M
Context
32K
Downloads
6.3M
Likes
689
Architecture
qwen2
24 layers · 2 KV heads · head 64 · hidden 896
GPU Requirements
VRAM / RAM / disk estimates by quantization and usage scenario
| Quantization | Minimum | Recommended | Production |
|---|---|---|---|
| FP1616bit | 6.1 GB RAM 32.0 · Disk 40.5 | 21.9 GB RAM 32.9 · Disk 66.5 | 118.9 GB RAM 178.3 · Disk 105.5 |
| FP88bit | 5.5 GB RAM 32.0 · Disk 39.7 | 21.3 GB RAM 32.0 · Disk 65.7 | 118.2 GB RAM 177.4 · Disk 104.7 |
| INT88bit | 5.5 GB RAM 32.0 · Disk 39.7 | 21.3 GB RAM 32.0 · Disk 65.7 | 118.2 GB RAM 177.4 · Disk 104.7 |
| AWQ4bit | 5.3 GB RAM 32.0 · Disk 39.4 | 21.0 GB RAM 32.0 · Disk 65.4 | 117.9 GB RAM 176.9 · Disk 104.4 |
| GPTQ4bit | 5.3 GB RAM 32.0 · Disk 39.4 | 21.0 GB RAM 32.0 · Disk 65.4 | 117.9 GB RAM 176.9 · Disk 104.4 |
| GGUF4bit | 5.3 GB RAM 32.0 · Disk 39.4 | 21.0 GB RAM 32.0 · Disk 65.4 | 118.0 GB RAM 176.9 · Disk 104.4 |
GPU Recommendations
1× RTX 5090
32 GB VRAM
≈ 7466.7 tok/s
Effective production throughput ≈ 5226.7 tok/s
score 0.6747
2× Tesla V100
32 GB VRAM
≈ 0 tok/s (N/A — benchmark unavailable)
Effective production throughput ≈ 0 tok/s
score 0.6212
8× H200 SXM
1128 GB VRAM
≈ 64000 tok/s
Effective production throughput ≈ 44800 tok/s
score 0.2197
1× L4
24 GB VRAM
≈ 1250 tok/s
Effective production throughput ≈ 875 tok/s
score 0.657
💬 Community — real deployment experience
0 Articles · 0 Benchmarks
GPUs that fit (single card)
L4
24 GB VRAM
$76.2
/mo · vast
RTX 3090
24 GB VRAM
$79.6
/mo · vast
RTX 5090
32 GB VRAM
$155
/mo · vast
A10
24 GB VRAM
$177
/mo · vast
RTX 4090
24 GB VRAM
$190
/mo · tensordock