Thread

The work by Valve's Timur Kristóf on improving old AMD GPUs on Linux

472 points · 97 comments · speckx

segmondy · 1791067001 · focus · HN ↗
I wonder if some of these could carry over for LLM inference. It will be nice to turn more ewaste GPUs into capable processing units for LLM.
1. lp92 · 1791079441 · focus · HN ↗
  Memory is the bottleneck along with the lack of FP4/FP8 capability at the hardware level.

Unofficial Hacker News client; not affiliated with Y Combinator.