Repository navigation
Extend MLX backend coverage to ModernBERT and Longformer architectures - #2455
Merged
Merged
Conversation
Owner
Author
|
Maintainer refresh on exact head 6393d30 against master 7c93a8e: the complete Metal-backed MLX suite passed with 166 tests and 1 expected skip; the 42 focused conversion, parity, and FP/INT8 tests passed with no skips; mypy and scoped hooks passed. This branch now consolidates #2340 and #2345; #2481 was closed as superseded. Hosted exact-head checks are starting. No merge performed. |
Owner
Author
|
Hosted Apple Silicon integration exposed a documentation-coherence failure: the newly supported ModernBERT and Longformer aliases were missing from docs/export-matrix.md. Repaired on exact head 3917459. The focused export-matrix test passes (3 tests), scoped hooks pass, and the strict documentation build passes. Exact-head CI is restarting. |
…r2455-modernbert-longformer
This was referenced Sep 30, 2026
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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
Adds native MLX token-classification backends for ModernBERT and Longformer, including architecture-specific normalization, converted-weight remapping, rotary/local attention, Longformer global attention, artifact loading, and INT8 shape preservation.
Type of Change
Testing
Validated at exact head
17cd2075ad72d11cb0ac043552035ea949a3c11dagainst master2c8c7b13ab757bd205a09899df0d33ca7aa2faa4on Apple Metal:Documentation
The MLX export support matrix and public family-resolution errors now include both backends and every supported alias.
Dependencies
No new dependency; tests use the existing optional MLX, PyTorch, and Transformers extras.
Related Issues
Closes #2340
Closes #2345
Screenshots/Examples
Not applicable; parity is covered by deterministic tiny reference models.