training diffusion step = 50？ #50

Open

opened

on Apr 30, 2025

Hello, I'd like to know what is your best training and inference diff_steps, thank you!
I saw in DiT, the default n_diff_steps = 500, is that for train or for inference?

to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Assignees

No one assigned

Labels

No labels

No labels

Type

No type

Projects

No projects

Milestone

No milestone

Relationships

None yet

Development

No branches or pull requests