Hello Folks,
I am looking for a tutorial or any experience on using NPUs in those Ryzen AI 300 series CPUs.
Does anybody know how to write code to utilize these NPUs?
Any Linux application that can make use of them?
I guess there is no kernel module to expose them, right?
Any link will be appreciated.
Please accept me at the beginner level.
Cheers.
Months later, has anyone documented use on Framework specifically?
Looks like some have used the extra abilities… Reddit - Please wait for verification
I am running fastflowlm (linked by Mario) on both my strix halo machines, Fedora 43, fastflow running in podman containers.
Had to build xdna from source and create a dkms for the driver but all working well alongside my vulcan-radv and rocm-7.2 llama.cpp containers.
Feels good to finally squeeze everything out of these little boxes.
Speaking of sqeezing, just saw that the the lemonade team has ported MLX to gfx1151
1 Like