from
Hacker News
Top
New
Run Deepseek from fast NVMe drives
by
ironbound
on 2/8/25, 2:13 PM with 3 comments
by
ironbound
on 2/8/25, 2:13 PM
Testing extreme NVME offload (4 x Gen5x4) for DeepSeek R1Because PCI-E 5x16 (~60GB/s) is close to dual channel DDR5 bandwidth, this is the cheapest method to run huge models. Code:
https://github.com/BlinkDL/fast.c