Does Magi-1 support batch inference? #120

Open

opened

on Dec 17, 2025

Does Magi-1 support batch inference?
I notice for one single output just use 10% of my GPU memory. Can I do batch inference to increase the generation efficiency?

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