[CP][RFC] Enable FlexCP for llama3 with parallelize_module #1707
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
Similar to #1696, but this PR uses parallel_module similar to TP/SP. This PR also requires pytorch/pytorch#162542
meta-cla
bot
added
the
CLA Signed
tianyu-l
reviewed
Sep 12, 2025
| device_mesh=world_mesh["cp"], | ||
| parallelize_plan=_ContextParallel( | ||
| seq_dim=2, | ||
| attention_type=_ContextParallel.AttentionType.FLEX, |
There was a problem hiding this comment.
Does this only work for FlexAttention?
Is there a plan to consolidate SDPA and FlexAttention in terms of how CP is applied?
There was a problem hiding this comment.
This will work for both SDPA and Flex. We just need to pass in a different type based on what attention is used.
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Outdated suggestions cannot be applied.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.
Similar to #1696, but this PR uses parallel_module similar to TP/SP.
This PR also requires pytorch/pytorch#162542