NVIDIA / NVIDIA/TensorRT-LLM

[Bug]: Whisper (PyTorch backend) fails with RuntimeError: bad optional access in the capacity scheduler — TensorRT-LLM 1.3.0rc25, RTX 5070 Ti (sm_120)

Open
#18,609 2 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

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

Description

System Info

X86
Sm120
Whisper large v3

Who can help?

No response

Information
  • The official example scripts
  • My own modified scripts
Tasks
  • An officially supported task in the examples folder (such as GLUE/SQuAD, ...)
  • My own task or dataset (give details below)
Reproduction

[Unable to encode/decode ](scheduler.py:462 schedule_request -> self.impl(...) -> RuntimeError: bad optional access)

Expected behavior

Active endpoint

actual behavior

HTTP 404

additional notes

Whisper support missing

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.

Research direction

Start at scheduler.py:462 and reproduce the failure with TensorRT-LLM 1.3.0rc25, Whisper large v3, and an sm_120 GPU. Trace the capacity-scheduler path behind RuntimeError: bad optional access and verify the reported endpoint behavior. Done means Whisper requests encode/decode successfully without the scheduler error and the endpoint no longer returns HTTP 404.

Written by the indexing model from the issue text.

Assessment

Tech stack
python, pytorch
Domain
ai, backend
Issue type
Bug
Difficulty
4/5
Estimated time
3-5 days
Activity status
Active
Clarity
Needs clarification
Newbie friendliness
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.