For local inference these days it’s basically all AS since it’s the only consumer hardware that can replace $20-80k server builds, but if you mean ML/AI developers specifically, they still prefer discrete GPUs for local training (on personal budgets: either recent consumer gaming/professional cards or older AI cards).
My bad, Apple Silicon, the private local ML “meta” as of ~Nov’25 (and last I saw) was RDMA clusters of M3 Ultras that could scale up to 3.6TB VRAM (a lot)
For local inference these days it’s basically all AS since it’s the only consumer hardware that can replace $20-80k server builds, but if you mean ML/AI developers specifically, they still prefer discrete GPUs for local training (on personal budgets: either recent consumer gaming/professional cards or older AI cards).
What does “AS” abbreviate here?
My bad, Apple Silicon, the private local ML “meta” as of ~Nov’25 (and last I saw) was RDMA clusters of M3 Ultras that could scale up to 3.6TB VRAM (a lot)