Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

No Improvement with DiTFastattn for opensora or video models #428

Open
nighting0le01 opened this issue Jan 12, 2025 · 2 comments
Open

No Improvement with DiTFastattn for opensora or video models #428

nighting0le01 opened this issue Jan 12, 2025 · 2 comments

Comments

@nighting0le01
Copy link

hi what thresholds and window sizes and settings to use for opensora and other video models when using DiTFastattn? i see no improvements whatsoever as it selects full_attn when using :https://github.com/thu-nics/DiTFastAttn/blob/main/experiments/opensora.sh.
please suggest
@feifeibear
Copy link
Collaborator

Could you post some performance data to illustrate this issue? What type of GPU did you use?

@feifeibear
Copy link
Collaborator

Additionally, it is important to clarify that we are not affiliated with OpenSora or DiTFastAttn. xDiT is a framework that incorporates DiTFastAttn.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
2 participants
@feifeibear @nighting0le01