Shatur@lemmy.ml to Linux@lemmy.ml · 11 months agoAMD Quietly Funded A Drop-In CUDA Implementation Built On ROCm: It's Now Open-Sourcewww.phoronix.comexternal-linkmessage-square20fedilinkarrow-up1503arrow-down15cross-posted to: linux@linux.communitystable_diffusion@lemmy.dbzer0.comprogramming@programming.devtechnology@lemmy.worldlinux@lemmy.worldtechnology@lemmy.worldopensource@lemmy.mlhackernews@lemmy.smeargle.fans
arrow-up1498arrow-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 · 11 months agomessage-square20fedilinkcross-posted to: linux@linux.communitystable_diffusion@lemmy.dbzer0.comprogramming@programming.devtechnology@lemmy.worldlinux@lemmy.worldtechnology@lemmy.worldopensource@lemmy.mlhackernews@lemmy.smeargle.fans
minus-squareMalReynolds@slrpnk.netlinkfedilinkEnglisharrow-up10·edit-210 months 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.