Commit History

Withdraw the MLX variant, Core ML is 2.7x faster
fb2096c
verified

msluszniak commited on

Stop restating quantized, and drop the unused default
2e4f31b

aceOfDiamonds commited on

Flatten preview/ away
45a7895

aceOfDiamonds commited on

Remove mlx export uploaded at incorrect top-level path (belongs under preview/)
107a4c4
verified

msluszniak commited on

Add mlx export (ExecuTorch 1.4.1)
b4bf3eb
verified

msluszniak commited on