Repository navigation
feat(cutile): add NVFP4 MoE batch-size benchmark - #1553
Merged
Merged
Conversation
Signed-off-by: Melih Elibol <elibol@users.noreply.github.com>
elibol
marked this pull request as ready for review
October 19, 2026 21:55
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Adds an NVFP4 MoE benchmark that validates outputs and measures CUDA graph replay latency for power-of-two batch sizes from 1 through 256 using synthetic inputs. Performance optimization is planned for a future version.
Validated on Blackwell with CUDA 13.3: full batch sweep, formatting, clippy, CPU/GPU suites, 27 supported examples, and eight benchmark smoke tests (smoke-test flag corrected locally in the runner).