Skip to content

[Bug]: Qwen Image sampling VRAM regression #1371

@dxqb

Description

@dxqb

What happened?

The Qwen VAE takes too much VRAM on torch 2.9.1 for 16 GB vram: pytorch/pytorch#177406
This is is already fixed in torch 2.10, but we are currently not upgrading because of pytorch/pytorch#175058

Workaround: Upgrade to torch 2.10 or downgrade to torch 2.8

What did you expect would happen?

Relevant log output

Generate and upload debug_report.log

No response

Metadata

Metadata

Assignees

No one assigned

    Labels

    bugSomething isn't workingwaitingwaiting for something / cannot fix currently

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions