NVIDIA / NVIDIA/TensorRT-LLM

[Feature]: AutoDeploy: MultiLorA Support

Open
#13,521 0 comments 0 reactions 1 assignee View on GitHub

@laikhtewari is already working on this.

Since Apr 27, 2026.

AutoDeploy feature request
Dominant language
Python
Stars
14.7k
Forks
2.8k
Avg merge
2d 23h
Merged PRs (30d)
489

Description

🚀 The feature, motivation and pitch

TP == 1
Correctness only - should work with MultiLorA.
No performance optimization passes (e.g. LorA Q/K/V fusions)
CUDA graph should be supported.
Can assume list of adapters is static if it helps - e.g., sequences come labeled with adapters (a new "slot") and then we can look up adapter info in a specific place.

Then later: generalize to a large list of adapters only a few of which are resident on GPU at a time.

Alternatives

No response

Additional context

No response

Before submitting a new issue...
  • Make sure you already searched for relevant issues, and checked the documentation and examples for answers to frequently asked questions.

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Assessment

This issue has not been assessed yet.

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.