Whisper: save converted weights as weights.safetensors - #1438
Open
sfeng01-tech wants to merge 1 commit into
Open
Whisper: save converted weights as weights.safetensors#1438sfeng01-tech wants to merge 1 commit into
sfeng01-tech wants to merge 1 commit into
Conversation
sfeng01-tech
marked this pull request as ready for review
August 18, 2026 01:50
sfeng01-tech
force-pushed
the
whisper-safetensors-compat
branch
from
August 18, 2026 01:53
b4328ae to
3248c3f
Compare
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Fixes #1421.
whisper/convert.pywrote converted weights asmodel.safetensors, but the releasedmlx-whisper==0.4.3loader expectsweights.safetensors(or the legacyweights.npz). Consequently, artifacts produced by the documented conversion command could not be loaded by the published package.This changes the converter to emit
weights.safetensors, which is compatible with both the released package and currentmain. It also adds a regression test that saves converted weights, verifies the artifact name and parameter keys, and reloads the model throughmlx_whisper.load_model. The Whisper conversion documentation now reflects the emitted artifact.Validation:
mlx-whisper==0.4.3: passed.git diff --check: passed.