Roboflow + Raspberry Pi
Run the Roboflow Inference server in Docker on a Raspberry Pi 4 or 5 and serve your models and Workflows from the device.
How to connect Raspberry Pi to Roboflow
5 stepsFlash a 64-bit OS
Install a 64-bit operating system on the Pi. Our docs suggest the 64-bit Raspberry Pi OS with desktop and recommended software if your SD card has room.
Install Docker
Install Docker on the Pi; the inference server runs as a container.
Start the inference server
Run pip install inference-cli, then inference server start. You can also run the roboflow/roboflow-inference-server-cpu image with docker run or Docker Compose.
Send images to the server
Point your code at http://localhost:9001 and run your model or Workflow on images from the Pi's camera or disk.
Open it to your network if needed
The server only answers on the device by default. Add --bind-address 0.0.0.0 to reach it over the LAN, after reviewing security, since the server has no authentication by default.
What teams build with Raspberry Pi
Watch a scene with a small camera
Run a compact detection model on a Pi next to the camera for monitoring that does not need high frame rates.
Prototype an edge deployment cheaply
Test a model and Workflow on low-cost hardware before moving to an NVIDIA Jetson for more speed.
Questions
The Raspberry Pi 4 Model B and Raspberry Pi 5, running a 64-bit operating system.
Inference does not use hardware acceleration on the Pi yet. Our docs estimate about 1 FPS on a Pi 4 and about 4 FPS on a Pi 5 for a Roboflow 3.0 Fast object detection model.
They will likely struggle on a Pi. For larger models or higher frame rates, we recommend an NVIDIA Jetson.
Yes, start it with --bind-address 0.0.0.0. Review security first, because the server has no authentication by default.
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.
