Enable direct MX4→BF16 dequantization to reduce memory (python side) (2/2)#5250
Closed
armandsauzay wants to merge 1 commit into
Closed
Enable direct MX4→BF16 dequantization to reduce memory (python side) (2/2)#5250armandsauzay wants to merge 1 commit into
armandsauzay wants to merge 1 commit into
Conversation
Contributor
|
@armandsauzay has exported this pull request. If you are a Meta employee, you can view the originating Diff in D88913122. |
…(2/2) (pytorch#5250) Summary: X-link: meta-pytorch/torchrec#3620 X-link: facebookresearch/FBGEMM#2241 python side of changes for Enabling direct MX4→BF16 dequantization to reduce memory Differential Revision: D88913122
armandsauzay
force-pushed
the
export-D88913122
branch
from
5055f5e to
98d7794
Compare
armandsauzay
pushed a commit
to armandsauzay/torchrec
that referenced
this pull request
Jan 6, 2026
…(2/2) (meta-pytorch#3620) Summary: X-link: pytorch/FBGEMM#5250 X-link: facebookresearch/FBGEMM#2241 python side of changes for Enabling direct MX4→BF16 dequantization to reduce memory Differential Revision: D88913122
meta-codesync Bot
pushed a commit
to meta-pytorch/torchrec
that referenced
this pull request
Jan 7, 2026
…(2/2) (#3620) Summary: X-link: pytorch/FBGEMM#5250 Pull Request resolved: #3620 X-link: https://github.com/facebookresearch/FBGEMM/pull/2241 python side of changes for Enabling direct MX4→BF16 dequantization to reduce memory Reviewed By: jwfromm, spcyppt Differential Revision: D88913122 fbshipit-source-id: 9d0c774e83e9bb5ccba7281b220f9ed54f690e5d
Contributor
|
This pull request has been merged in 562677c. |
jensenhwa
pushed a commit
to jensenhwa/FBGEMM
that referenced
this pull request
Mar 18, 2026
…(2/2) (pytorch#5250) Summary: Pull Request resolved: pytorch#5250 X-link: meta-pytorch/torchrec#3620 X-link: https://github.com/facebookresearch/FBGEMM/pull/2241 python side of changes for Enabling direct MX4→BF16 dequantization to reduce memory Reviewed By: jwfromm, spcyppt Differential Revision: D88913122 fbshipit-source-id: 9d0c774e83e9bb5ccba7281b220f9ed54f690e5d
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
3 participants
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.
Summary:
X-link: meta-pytorch/torchrec#3620
X-link: https://github.com/facebookresearch/FBGEMM/pull/2241
python side of changes for Enabling direct MX4→BF16 dequantization to reduce memory
Differential Revision: D88913122