Stay organized with collectionsSave and categorize content based on your preferences.
The following codelab shows how to run a backend service that runsvLLM, which is an
inference engine for production systems, along with Google'sGemma 2, which is
a 2 billion parameters instruction-tuned model.
[[["Easy to understand","easyToUnderstand","thumb-up"],["Solved my problem","solvedMyProblem","thumb-up"],["Other","otherUp","thumb-up"]],[["Hard to understand","hardToUnderstand","thumb-down"],["Incorrect information or sample code","incorrectInformationOrSampleCode","thumb-down"],["Missing the information/samples I need","missingTheInformationSamplesINeed","thumb-down"],["Other","otherDown","thumb-down"]],["Last updated 2025-09-04 UTC."],[],[],null,["# Run LLM inference on Cloud Run GPUs with vLLM\n\nThe following codelab shows how to run a backend service that runs [vLLM](https://github.com/vllm-project/vllm), which is an\ninference engine for production systems, along with Google's [Gemma 2](https://developers.googleblog.com/en/smaller-safer-more-transparent-advancing-responsible-ai-with-gemma/), which is\na 2 billion parameters instruction-tuned model.\n\nSee the entire codelab at [Run LLM inference on Cloud Run GPUs with vLLM](https://codelabs.developers.google.com/codelabs/how-to-run-inference-cloud-run-gpu-vllm#0)."]]
