DeepSeek R1 Distill Qwen 7B
deepseek-ai/DeepSeek-R1-Distill-Qwen-7B
Parameters
7.6B
Context
128K
Downloads
250.4K
Likes
870
Architecture
transformer
GPU Requirements
VRAM / RAM / disk estimates by quantization and usage scenario
| Quantization | Minimum | Recommended | Production |
|---|---|---|---|
| FP1616bit | 22.9 GB RAM 34.3 · Disk 61.1 | 39.4 GB RAM 59.1 · Disk 87.1 | 137.1 GB RAM 205.7 · Disk 126.1 |
| FP88bit | 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 |
| INT88bit | 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 |
| AWQ4bit | 9.6 GB RAM 32.0 · Disk 44.7 | 25.6 GB RAM 38.3 · Disk 70.7 | 122.7 GB RAM 184.0 · Disk 109.7 |
| GPTQ4bit | 9.7 GB RAM 32.0 · Disk 44.8 | 25.6 GB RAM 38.4 · Disk 70.8 | 122.7 GB RAM 184.1 · Disk 109.8 |
| GGUF4bit | 9.8 GB RAM 32.0 · Disk 44.9 | 25.7 GB RAM 38.6 · Disk 70.9 | 122.9 GB RAM 184.3 · Disk 109.9 |
GPU Recommendations
1× RTX 5090
32 GB VRAM
≈ 488.3 tok/s
Effective production throughput ≈ 341.8 tok/s
score 0.6847
2× Tesla V100
32 GB VRAM
≈ 0 tok/s (N/A — benchmark unavailable)
Effective production throughput ≈ 0 tok/s
score 0.6336
8× H200 SXM
1128 GB VRAM
≈ 4185.3 tok/s
Effective production throughput ≈ 2929.7 tok/s
score 0.2197
1× RTX A6000
48 GB VRAM
≈ 209.3 tok/s
Effective production throughput ≈ 146.5 tok/s
score 0.6331
💬 Community — real deployment experience
0 Articles · 0 Benchmarks
GPUs that fit (single card)
RTX A6000
48 GB VRAM
$210
/mo · vast
RTX 5090
32 GB VRAM
$246
/mo · vast
L40S
48 GB VRAM
$341
/mo · vast
RTX PRO 6000 S
48 GB VRAM
$633
/mo · vast
RTX PRO 6000 WS
48 GB VRAM
$634
/mo · vast