Qwen/Qwen2.5-VL-7B-Instruct
Qwen/Qwen2.5-VL-7B-Instruct
Qwen2.5-VL 7B, vision-language with strong OCR.
Parameters
5.3B
Context
125K
Downloads
5.3M
Likes
1.7K
Architecture
qwen2_5_vl
28 layers · 4 KV heads · head 128 · hidden 3584
GPU Requirements
VRAM / RAM / disk estimates by quantization and usage scenario
| Quantization | Minimum | Recommended | Production |
|---|---|---|---|
| FP1616bit | 24.5 GB RAM 36.8 · Disk 63.1 | 41.2 GB RAM 61.7 · Disk 89.1 | 138.9 GB RAM 208.4 · Disk 128.1 |
| FP88bit | 14.7 GB RAM 32.0 · Disk 51.1 | 30.9 GB RAM 46.4 · Disk 77.1 | 128.3 GB RAM 192.4 · Disk 116.1 |
| INT88bit | 14.7 GB RAM 32.0 · Disk 51.1 | 30.9 GB RAM 46.4 · Disk 77.1 | 128.3 GB RAM 192.4 · Disk 116.1 |
| AWQ4bit | 10.0 GB RAM 32.0 · Disk 45.3 | 26.0 GB RAM 39.0 · Disk 71.3 | 123.1 GB RAM 184.7 · Disk 110.3 |
| GPTQ4bit | 10.1 GB RAM 32.0 · Disk 45.3 | 26.1 GB RAM 39.1 · Disk 71.3 | 123.2 GB RAM 184.8 · Disk 110.3 |
| GGUF4bit | 10.2 GB RAM 32.0 · Disk 45.5 | 26.2 GB RAM 39.3 · Disk 71.5 | 123.3 GB RAM 185.0 · Disk 110.5 |
GPU Recommendations
1× H100 SXM
80 GB VRAM
≈ 835.4 tok/s
Effective production throughput ≈ 584.8 tok/s
score 0.56
2× Tesla V100
32 GB VRAM
≈ 0 tok/s (N/A — benchmark unavailable)
Effective production throughput ≈ 0 tok/s
score 0.6293
8× H200 SXM
1128 GB VRAM
≈ 3830.4 tok/s
Effective production throughput ≈ 2681.3 tok/s
score 0.2197
1× A100 SXM4
80 GB VRAM
≈ 0 tok/s (N/A — benchmark unavailable)
Effective production throughput ≈ 0 tok/s
score 0.572
💬 Community — real deployment experience
0 Articles · 0 Benchmarks
GPUs that fit (single card)
A100 SXM4
80 GB VRAM
$225
/mo · vast
RTX 5090
32 GB VRAM
$302
/mo · vast
L40S
48 GB VRAM
$322
/mo · vast
RTX A6000
48 GB VRAM
$351
/mo · vast
A100 PCIE
80 GB VRAM
$354
/mo · vast