feat(quantization): support MOE_REQUANTIZE_WEIGHT_DTYPE and MOE_REQUANTIZE_BLOCK_SIZE in MXFP4 MoE - #3494
Open
aashishrampal-lab wants to merge 1 commit into
Conversation
…NTIZE_BLOCK_SIZE in MXFP4 MoE
aashishrampal-lab
requested review from
QiliangCui,
gxd3,
jrplatin,
kyuyeunk,
vanbasten23 and
vipannalla
as code owners
August 31, 2026 09:55
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Supports configurable requantization dtype (e.g. FP8) and block size for MXFP4 MoE weights.
Part 1 of the DeepSeek-V4 Pipeline Parallelism & Layerwise Loading stack.