Feature/int4 lut gemm #32
Merged
Conversation
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Outdated suggestions cannot be applied.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.
This update refactors the lookup table implementation to support mixed-precision matrix multiplication, specifically INT4 × FP16 or FP32. The
ProductLookupTableclass has been extended to allow both scalar lookups and dynamic row-wise generation of lookup vectors based on activation matrix rows. This enables more efficient LUT-based computation without requiring runtime multiplications. The fast GEMM kernel is updated accordingly to rebuild the LUT for each activation row, improving cache locality and performance. Additionally, the original AVX2-specific code has been removed to simplify the implementation and improve readability. All relevant tests, including those involving INT4 and mixed-precision computation, pass successfully.