feat: enable head_dim=256 for attention kernels
#132
Merged
Conversation
This file contains 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
This was referenced Feb 25, 2024
yzh119
added a commit
that referenced
this pull request
Feb 25, 2024
3 tasks
yzh119
added a commit
that referenced
this pull request
Mar 8, 2024
🤖 I have created a release *beep* *boop* --- ## [0.0.3](v0.0.3...v0.1.0) (2024-03-08) ### Features * adding `sm_scale` field for all attention APIs ([#145](#145)) ([85d4018](85d4018)) * enable `head_dim=256` for attention kernels ([#132](#132)) ([0372acc](0372acc)) * pytorch api of fp8 kv-cache ([#156](#156)) ([66ee066](66ee066)) * support ALiBi ([#146](#146)) ([383518b](383518b)) ### Misc * add stream argument in BeginForwardFunction of TVMWrapper ([#164](#164)) ([fabfcb5](https://github.com/flashinfer-ai/flashinfer/tree/fabfcb5751dcc003137a5a7d2d5514f3afe2e302)) ### Bug Fixes * bugfix to pr 135 ([#136](#136)) ([3d55c71](3d55c71)) * fix bugs introduced in [#132](#132) ([#135](#135)) ([9b7b0b9](9b7b0b9)) * fix FindThrust.cmake ([#161](#161)) ([30fa584](30fa584)) ### Performance Improvements * multiple q by sm_scale in decode kernels ([#144](#144)) ([660c559](660c559)) --- This PR was generated with [Release Please](https://github.com/googleapis/release-please). See [documentation](https://github.com/googleapis/release-please#release-please). --------- Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com> Co-authored-by: yzh119 <expye@outlook.com>
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Outdated suggestions cannot be applied.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.
As mentioned in #130 , the kernels for
head_dim=256are not compiled by default, this PR expose these attention kernels to pip wheels and adds unittests/benchmarks forhead_dim=256.