Skip to content

Rapid model loading #18

Closed as not planned
Feature
Closed as not planned
Rapid model loading#18
Feature
Labels
inference-stack

Description

@fergusfinn
fergusfinn
opened on Sep 8, 2025

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

Stub, will expand

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

Metadata

Metadata

Assignees

No one assigned

    Labels

    inference-stack

    Type

    Feature

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions