Rapid model loading #18

Closed as not planned

Labels

inference-stack

opened

Use NIXL to fast load models into GPU memory from a mirror.

Stub, will expand

Metadata

Assignees

No one assigned

Labels

inference-stack

Type

Projects

No projects

Milestone

No milestone

Relationships

None yet

Development

No branches or pull requests