optimize ace_step: completely remove torch runtime dependency + unified conversion pipeline#577
Merged
Merged
Conversation
mm65x
changed the title
mm65x
changed the title
Collaborator
|
@mm65x Any change you can put the converted model in |
lucasnewman
reviewed
Mar 19, 2026
mm65x
force-pushed
the
pc/enhance-ace-step
branch
from
March 20, 2026 10:14
d081433 to
1386580
Compare
Contributor
Author
|
done |
Contributor
Author
|
uploaded the converted model here: https://huggingface.co/mlx-community/ACE-Step-v1.5-turbo-mlx repo id is |
Owner
|
Please name it by dtype or quant type
Or
The convert script already adds tags that identify the model as MLX 👌🏽 |
Contributor
Author
|
done |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Description
this PR builds directly on the massive native MLX implementation from #499, optimizing the
ace_stepimplementation by completely removing the need for PyTorch at runtime, and adding a robust unified conversion script.i noticed two small things missing from the branch that this PR addresses:
ace_stepgeneration module still fell back toimport torchjust to load thesilence_latent.pttensor, which technically means users still need a massive PyTorch installation just to run inference.convert.pyscript for users to pull the raw HuggingFace weights and quantize/package them into a neatsafetensorsdirectory format locally.Changes in the codebase
convert.pythat hooks directly into the upstream HuggingFace repository to extract the DiT, the VAE, and the Text Encoder.silence_latent.ptinto a puresilence_latent.npynumpy array.import torchruntime dependency inace_step.pycompletely, hooking it up to read thenpyfile vianumpyinstead!this makes the generation pipeline 100% native PyTorch-free at runtime!