Roboflow + Google Cloud Run
Roboflow Inference ships as a Docker image, so you can run it on Google Cloud Run like any other container. Current Roboflow docs do not include a Cloud Run guide.
How to connect Google Cloud Run to Roboflow
3 stepsMirror the Inference image
Create an Artifact Registry remote repository for Docker Hub, as Google recommends, or reference roboflow/roboflow-inference-server-cpu on Docker Hub directly.
Create a Cloud Run service
Deploy a new service from that image and set the container port to 9001.
Call the service from your app
Send requests to the service URL with your Roboflow API key, using it as the `api_url` in the Python SDK.
What teams build with Google Cloud Run
Serve occasional image requests
Run a CPU model behind an HTTP endpoint inside your Google Cloud project.
Keep images inside Google Cloud
Process images in your own project instead of sending them to Roboflow's hosted API.
Questions
Not currently. Roboflow's Google Cloud guide covers Compute Engine VMs set up with the Inference CLI, and the Linux guide covers running the Docker image yourself.
Cloud Run can deploy public Docker Hub images, but Google recommends pulling them through an Artifact Registry remote repository.
If you do not want to manage infrastructure, Roboflow's Serverless Hosted API and Dedicated Deployments run models for you.
Why are companies choosing Roboflow?
The end-to-end platform for vision AI: create datasets, train models and deploy to production.
Flexible deployment modes
Move from a scalable cloud API to edge hardware like NVIDIA Jetson without managing complex dependencies.
Integrated workflow builder
Combine multiple AI models, custom business logic and third-party integrations into one production-ready pipeline in a low-code editor.
Model monitoring and alerts
Dashboards track confidence, inference time and request volume for your deployed models, and email alerts tell your team when confidence suddenly drops.
