You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Hi @raviteja5 , M1 is supported in cpu mode, for distilgpt2_lora model it should work good (~270Mb), you can also use gpt2_lora (~1GB) and gpt2 (~3.5GB)
We are also doing experiments to support M1 gpu mode to get speedup, but it is working only in pytorch dev build
Thanks for your efforts in making this library. I have a M1 mac, and it says CUDA is not available.
Do you support acceleration on Mac with Metal (MPS) etc.?
Thanks!
The text was updated successfully, but these errors were encountered: