Add example: deploy Faster-Whisper as a custom model on a SageMaker Triton MME#4887
Add example: deploy Faster-Whisper as a custom model on a SageMaker Triton MME#4887eldadcohen1 wants to merge 12 commits into
Conversation
…ased on faster whisper
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
|
Check out this pull request on See visual diffs & provide feedback on Jupyter Notebooks. Powered by ReviewNB |
|
@mollyheamazon can you please take a look |
|
Hi @atqy — could a maintainer help get the CI moving on this PR? The three required AWS CodeBuild us-west-2 (SageMakerNotebooksProd-USWEST2-...) checks are stuck as "expected" and have never started. I verified via the GitHub API that the head commit ( Some context:
Could you please re-trigger the CodeBuild checks for this PR (or let me know if the fork-PR CI pipeline is currently inactive and how you'd like me to proceed)? Happy to push an empty commit or close/reopen if that helps re-fire the webhook. Thanks! |
|
Superseded by #4888, which targets the current default branch and ports the example into the new repository structure ( |
Use Faster-Whisper as a custom model artifact from Hugging Face that is served on Amazon SageMaker as a multi-model endpoint with Triton.
Description of changes:
Adds a new example notebook under
sagemaker-triton/mme-triton-custom-faster-whisper/showing how to package a pre-trained Faster-Whisper-Large-v3 model, build a custom Triton container, and deploy it to a SageMaker GPU multi-model endpoint. Described in the readme.This PR supersedes #4608, which had a stale branch and a head branch of
main. This one is rebased on the latest upstreammainand uses a dedicated feature branch.Testing is done:
Tested on AWS SageMaker.
Merge Checklist
black-nb -l 100 {path}/{notebook-name}.ipynbBy submitting this pull request, I confirm that my contribution is made under the terms of the Apache 2.0 license.