not_IO@lemmy.blahaj.zone to Programmer Humor@programming.devEnglish · 2 days agomight be a form of Jevons Paradoxlemmy.blahaj.zoneimagemessage-square256fedilinkarrow-up11.54Karrow-down111file-text
arrow-up11.53Karrow-down1imagemight be a form of Jevons Paradoxlemmy.blahaj.zonenot_IO@lemmy.blahaj.zone to Programmer Humor@programming.devEnglish · 2 days agomessage-square256fedilinkfile-text
minus-squareSamskara@sh.itjust.workslinkfedilinkarrow-up1·10 hours agoHave you ever tried running a decent sized LLM locally?
minus-squareVlyn@lemmy.ziplinkfedilinkEnglisharrow-up1·10 hours agoDecent sized for what? Creative writing and roleplay? Plenty, but I try to fit it into my 16 GB VRAM as otherwise it’s too slow for my liking. Coding/complex tasks? No, that would need 128GB and upwards and it would still be awfully slow. Except you use a Mac with unified memory. For image and video generation you’d want to fit it into GPU VRAM again, system RAM would be way too slow.
minus-squareSamskara@sh.itjust.workslinkfedilinkarrow-up1·9 hours agoI use a Mac with unified memory, so that distinction slipped my mind.
Have you ever tried running a decent sized LLM locally?
Decent sized for what?
Creative writing and roleplay? Plenty, but I try to fit it into my 16 GB VRAM as otherwise it’s too slow for my liking.
Coding/complex tasks? No, that would need 128GB and upwards and it would still be awfully slow. Except you use a Mac with unified memory.
For image and video generation you’d want to fit it into GPU VRAM again, system RAM would be way too slow.
I use a Mac with unified memory, so that distinction slipped my mind.