[CUDA] Heuristics for Hopper QMM - #3173
Merged
Merged
Conversation
zcbenz
force-pushed
the
qmm-sm90-dispatch
branch
6 times, most recently
from
February 26, 2026 23:18
1f49732 to
9c7b35f
Compare
zcbenz
force-pushed
the
qmm-sm90-dispatch
branch
from
February 26, 2026 23:36
9c7b35f to
1675962
Compare
jzdziarski
pushed a commit
to jzdziarski/mlx
that referenced
this pull request
Jul 26, 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
2 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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Refs #2536, #3160.
Add simple heuristics for QMM by dispatching to different tile shapes depending on the M. This does not change anything for small M problems and intends to fix bad performance for large M.
Also I realized that it is not possible to implement fp quantizations with the cutlass kernel because minimum group size is 64, so I simplified the checks in
qmm_sm90.