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

[feature] apply USP+FSDP to reduce memory overhead #400

Open
feifeibear opened this issue Dec 19, 2024 · 0 comments
Open

[feature] apply USP+FSDP to reduce memory overhead #400

feifeibear opened this issue Dec 19, 2024 · 0 comments

Comments

@feifeibear
Copy link
Collaborator

https://pytorch.org/docs/stable/fsdp.html

simply wrapper the model with FSDP. Each GPU maintains a shard of parameters and conduction parameter communication during inference.

The solution is easily compatible with USP.
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
1 participant
@feifeibear