RTX 5090 does not meet the minimum VRAM requirement for Q4 inference of Qwen/Qwen2.5-72B-Instruct. Review the quantization breakdown below to see how higher precision settings impact VRAM and throughput.
RTX 5090 lacks sufficient VRAM for comfortable Qwen/Qwen2.5-72B-Instruct operation with Q4 quantization.
Your 32GB GPU is 3GB short of the 35GB minimum.
Options: (1) Try Q2 or Q3 quantization for lower VRAM requirements, (2) Consider cloud GPU rental, (3) Upgrade to a GPU with at least 16GB VRAM.
| Quantization | VRAM needed | VRAM available | Estimated speed | Verdict |
|---|---|---|---|---|
| Q4 | 35GB | 32GB | 64.11 tok/s | ❌ Not recommended |
| Q8 | 70GB | 32GB | 38.32 tok/s | ❌ Not recommended |
| FP16 | 141GB | 32GB | 24.20 tok/s | ❌ Not recommended |