HN
Hacker News
Top
New
Best
Ask
Show
Jobs
Comment by vikmals | Hacker News Reader
Parent
Full thread
vikmals
·
fastllm targets the GPU, while colibri uses CPU inference only
View on HN
aliljet
·
I'd be curious about an.option that would allow glm use with a low end GPU like a 2080 ti...
Reply on news.ycombinator.com