[RELEASE] Fix compression of first & last layers of bitnet to int8 (#3738) #3739
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Outdated suggestions cannot be applied.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.
#3738 to the release branch
The fix is pretty-simple, but works with OpenVINO 2025.4
Once optimum is released, should work out-of-the box
Changes
Added u2 type support in NNCF graph
Reason for changes
int8 compression of first/last layer of bitnet fails with error:
Related tickets
CVS-176501
Tests
conformance weight compression on after #3738 merge: https://github.com/openvinotoolkit/nncf/actions/runs/19437926992
conformance quantization on after #3738 merge: manual/job/post_training_quantization/755/
manual, since need optimum from main branch and openvino==2025.4 and OV doesn't expose
ov.Type.u2in Python API (requested for 2026.0 release)