✅ — AWS EC2 P5 Instance (8x H100) Can Run Qwen 2.5 3B
The AWS EC2 P5 Instance (8x H100) packs 640GB VRAM. Qwen 2.5 3B needs ~1.7GB. You're good to go.
✅ Plenty of headroom
Hardware Specs
- GPU
- AWS EC2 P5 Instance (8x H100)
- VRAM
- 640 GB
- Bandwidth
- 26400 GB/s
- MSRP
- $0
Model Requirements
- Model
- Qwen 2.5 3B
- Parameters
- 3B
- VRAM (Q4)
- ~1.7 GB
- Context
- N/A
Estimated Performance
13200 tok/s
Estimated inference speed (Q4 quantization)