Shatur@lemmy.ml to Linux@lemmy.ml · 2 years agoAMD Quietly Funded A Drop-In CUDA Implementation Built On ROCm: It's Now Open-Sourcewww.phoronix.comexternal-linkmessage-square29linkfedilinkarrow-up1521arrow-down15cross-posted to: technology@lemmy.worldtechnology@lemmy.world
arrow-up1516arrow-down1external-linkAMD Quietly Funded A Drop-In CUDA Implementation Built On ROCm: It's Now Open-Sourcewww.phoronix.comShatur@lemmy.ml to Linux@lemmy.ml · 2 years agomessage-square29linkfedilinkcross-posted to: technology@lemmy.worldtechnology@lemmy.world
minus-squareMalReynolds@slrpnk.netlinkfedilinkEnglisharrow-up10·2 years agoYes, llama.cpp and derivates, stable diffusion, they also run on ROCm. LLM fine-tuning is CUDA as well, ROCm implementations not so much for this, but coming along.
Yes, llama.cpp and derivates, stable diffusion, they also run on ROCm. LLM fine-tuning is CUDA as well, ROCm implementations not so much for this, but coming along.