Skip to content

Conversation

@e-cal
Copy link

@e-cal e-cal commented Apr 23, 2025

Use huggingface builtin device_map to auto split the model. Enables users with less vram to still use the model. (e.g. on my 12GB gpu I go from OOM errors to running just fine).

@jaehong21 jaehong21 added the enhancement New feature or refactor label Apr 23, 2025
@e-cal
Copy link
Author

e-cal commented Apr 28, 2025

fixed conflicts

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

Labels

enhancement New feature or refactor

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants