The Bee Stack repository provides everything you need to run the Bee Application Stack locally using Docker Compose. This setup allows you to run, test, and experiment with Bee's various components seamlessly.
The Bee Stack comprises the following components, each contributing distinct functionalities to support your AI-driven applications:
- bee-agent-framework gives the foundation to build LLM Agents.
- bee-code-interpreter runs a user or generated Python code in a sandboxed environment.
- bee-api exposes agents via OpenAPI compatible Rest API.
- bee-ui allows you to create agents within your web browser.
- bee-observe and bee-observe-connector help you to trace what you are agents are doing.
Docker or similar container engine including docker compose (Rancher desktop or Podman)
⚠️ Warning: A rootless machine is not supported (e.g. if you use podman, set your VM to rootful)
- Clone this repository
git clone https://github.com/i-am-bee/bee-stack.git
cd bee-stack- Configure environment and fill required variables
cp example.env .env
vim .env # fill in your API key- Up! (this might take a while the first time you run it)
docker compose --profile all up -dOnce started you can find use the following URLs:
- bee-ui: http://localhost:3000
- mlflow: http://localhost:8080
- bee-api: http://localhost:4000 (for direct use of the api, use apiKey
sk-testkey) - list all open ports:
docker compose ps --format "{{.Names}}: {{.Ports}}"
You can use any typical compose commands to inspect the state of the services:
docker compose ps
docker compose logs bee-apiStopping services:
# Stop all
docker compose --profile all down
# Stop all and remove data
docker compose --profile all down --volumesIf you are a developer on bee-api or bee-ui and want to run only the supporting infrastructure,
use the profile infra, e.g.:
docker compose --profile infra up -dThe Bee Agent Framework is an open-source project and we ❤️ contributions.
If you'd like to contribute to Bee, please take a look at our contribution guidelines.
Special thanks to our contributors for helping us improve Bee Agent Framework.
