• 1 Post
  • 54 Comments
Joined 1 year ago
cake
Cake day: August 8th, 2023

help-circle









  • IDK, looks like 48GB cloud pricing would be 0.35/hr => $255/month. Used 3090s go for $700. Two 3090s would give you 48GB of VRAM, and cost $1400 (I’m assuming you can do “model-parallel” will Llama; never tried running an LLM, but it should be possible and work well). So, the break-even point would be <6 months. Hmm, but if Severless works well, that could be pretty cheap. Would probably take a few minutes to process and load a ~48GB model every cold start though?


  • 31337@sh.itjust.workstoMemes@lemmy.mlAmd fan
    link
    fedilink
    arrow-up
    3
    ·
    3 months ago

    I’ve put together 2 computers the last couple years, one Intel (12th gen, fortunately) and one AMD. Both had stability issues, and I had to mess with the BIOS settings to get them stable. I actually had to under-clock the RAM on the AMD (probably had something to do with maxing-out the RAM capacity, but I still shouldn’t need to under-clock, IMO). I think I’m going to get workstation-grade components the next time I need to build a computer.