Optimize fp8 headim64 tile #91
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
Signed-off-by: Jonas Kuebler <kuebj@amazon.com>
Signed-off-by: Jonas Kuebler <kuebj@amazon.com>
jmkuebler
force-pushed
the
optimize_fp8_headim64_tile
branch
from
c689a53 to
90a6e1d
Compare
1 task
LucasWilkinson
approved these changes
Sep 19, 2025
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Outdated suggestions cannot be applied.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.
This PR commits an optimized tile size to speedup decoding for models w/ head_dim 64 when running in FP8. It applies for example to GPT-OSS.
See vllm-project/vllm#24916 (optimization 3) for the improvement.
cc @LucasWilkinson