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
Creating a meta issue for clarity / visibility so others can upvote / the team can prioritize.
It would be extremely useful to have a Swift ml-explore SDK that can:
Be used in iOS / iPadOS applications
Run using the A-series and A- or M-series hardware acceleration (GPU+ANE).
One might not even need to code the model in Python mlx in the first place: popular open source architectures could be written directly in Swift and used by many developers.
The text was updated successfully, but these errors were encountered:
Getting models implemented in PyTorch to run on iOS using CoreML is a painful developer experience and it would be easier to just use a Swift mlx frontend for on-device training / inference.
Creating a meta issue for clarity / visibility so others can upvote / the team can prioritize.
It would be extremely useful to have a Swift ml-explore SDK that can:
One might not even need to code the model in Python
mlx
in the first place: popular open source architectures could be written directly in Swift and used by many developers.The text was updated successfully, but these errors were encountered: