Qwen/Qwen3.5-9B
Qwen/Qwen3.5-9B
Parameters
7.8B
Context
256K
Downloads
9.6M
Likes
2.1K
Architecture
qwen3_5
32 layers · 4 KV heads · head 256 · hidden 4096
GPU Requirements
VRAM / RAM / disk estimates by quantization and usage scenario
| Quantization | Minimum | Recommended | Production |
|---|---|---|---|
| FP1616bit | 26.2 GB RAM 39.2 · Disk 65.2 | 42.9 GB RAM 64.3 · Disk 91.2 | 140.7 GB RAM 211.1 · Disk 130.2 |
| FP88bit | 15.6 GB RAM 32.0 · Disk 52.1 | 31.8 GB RAM 47.7 · Disk 78.1 | 129.2 GB RAM 193.8 · Disk 117.1 |
| INT88bit | 15.6 GB RAM 32.0 · Disk 52.1 | 31.8 GB RAM 47.7 · Disk 78.1 | 129.2 GB RAM 193.8 · Disk 117.1 |
| AWQ4bit | 10.5 GB RAM 32.0 · Disk 45.8 | 26.5 GB RAM 39.7 · Disk 71.8 | 123.6 GB RAM 185.4 · Disk 110.8 |
| GPTQ4bit | 10.5 GB RAM 32.0 · Disk 45.9 | 26.5 GB RAM 39.8 · Disk 71.9 | 123.7 GB RAM 185.5 · Disk 110.9 |
| GGUF4bit | 10.7 GB RAM 32.0 · Disk 46.0 | 26.7 GB RAM 40.0 · Disk 72.0 | 123.8 GB RAM 185.7 · Disk 111.0 |
GPU Recommendations
1× H100 SXM
80 GB VRAM
≈ 770.1 tok/s
Effective production throughput ≈ 539.1 tok/s
score 0.5728
2× Tesla V100
32 GB VRAM
≈ 0 tok/s (N/A — benchmark unavailable)
Effective production throughput ≈ 0 tok/s
score 0.625
8× H200 SXM
1128 GB VRAM
≈ 3531 tok/s
Effective production throughput ≈ 2471.7 tok/s
score 0.2197
1× RTX A6000
48 GB VRAM
≈ 176.6 tok/s
Effective production throughput ≈ 123.6 tok/s
score 0.6369
💬 Community — real deployment experience
0 Articles · 0 Benchmarks
GPUs that fit (single card)
RTX A6000
48 GB VRAM
$264
/mo · vast
RTX 5090D
32 GB VRAM
$294
/mo · vast
RTX 5090
32 GB VRAM
$340
/mo · vast
L40S
48 GB VRAM
$341
/mo · vast
A100 SXM4
80 GB VRAM
$344
/mo · vast