Run LLM inference on Cloud Run GPUs with Hugging Face Transformers.js
Stay organized with collections
Save and categorize content based on your preferences.
Cloud Run GPUs let you run large language model (LLM) inference in a
fully managed environment. In this tutorial, you deploy a backend
service that uses the Transformers.js
package, the
JavaScript equivalent of the Hugging Face transformers
Python library, to serve Google's
Gemma 2
model.
To complete the step-by-step walkthrough, see the codelab for
How to Run Transformers.js on Cloud Run GPUs.
Except as otherwise noted, the content of this page is licensed under the Creative Commons Attribution 4.0 License, and code samples are licensed under the Apache 2.0 License. For details, see the Google Developers Site Policies. Java is a registered trademark of Oracle and/or its affiliates.
Last updated 2026-10-30 UTC.
[null,null,["Last updated 2026-10-30 UTC."],[],[]]
