Install on MacOS¶

OSX Native App (Apple Silicon)¶

You can now run Roboflow Inference Server on your Apple Silicon Mac using our native desktop app!

Simply download the latest DMS disk image from the latest release on Github. ➡️ View Latest Release and Download Installers on Github

OSX Installation Steps¶

Download the Roboflow Inference DMG disk image
Mount hte disk image by double clicking it
Drag the Roboflow Inference App to the Application Folder
Go to your Application Folder and double click the Roboflow Inference App to start the server

Using Docker¶

CPUGPU

First, you'll need to install Docker Desktop. Then, use the CLI to start the container.

pip install inference-cli
inference server start

Manually Starting the Container¶

If you want more control of the container settings you can also start it manually:

sudo docker run -d \
    --name inference-server \
    --read-only \
    -p 9001:9001 \
    --volume ~/.inference/cache:/tmp:rw \
    --security-opt="no-new-privileges" \
    --cap-drop="ALL" \
    --cap-add="NET_BIND_SERVICE" \
    roboflow/roboflow-inference-server-cpu:latest

Apple does not yet support passing the Metal Performance Shader (MPS) device to Docker so hardware acceleration is not possible inside a container on Mac.

Tip

It's easiest to get started with the CPU Docker and switch to running outside of Docker with MPS acceleration later if you need more speed.

We recommend using pyenv and pyenv-virtualenv to manage your Python environments on Mac (especially because, in 2025, homebrew is defaulting to Python 3.13 which is not yet compatible with several of the machine learning dependencies that Inference uses).

Once you have installed and setup pyenv and pyenv-virtualenv (be sure to follow the full instructions for setting up your shell), create and activate an inference virtual environment with Python 3.12:

pyenv install 3.12
pyenv virtualenv 3.12 inference
pyenv activate inference

To install and run the server outside of Docker, clone the repo, install the dependencies, copy cpu_http.py into the top level of the repo, and start the server with uvicorn:

git clone https://github.com/roboflow/inference.git
cd inference
pip install .
cp docker/config/cpu_http.py .
uvicorn cpu_http:app --port 9001 --host 0.0.0.0

Your server is now running at localhost:9001 with MPS acceleration.

Using Your New Server¶

See Using Your New Server for next steps.