Practical overview
When it fits
Use Burr when a Python application needs explicit state management, complex decision paths, human feedback or inspectable execution, and you want to manage its runtime infrastructure yourself.
Who it is for
- Python developers building chatbots or agents with stateful workflows.
- Teams that need to trace and inspect application decisions during execution.
- Developers building simulations or other stateful applications, including non-LLM use cases.
Documented uses
- Represent application steps and decision paths as a state machine.
- Build a chatbot or agent that reads and updates application state.
- Monitor and trace application execution in the included UI.
- Save and load application state using pluggable persisters.
- Connect an LLM, framework or other library to a workflow without relying on Burr to define model or data handling.
Requirements & limits
The project is open source under the Apache-2.0 license. The README documents installation from PyPI; it does not state a price.
- Applications are authored in Python and expressed as state machines using actions, state and transitions.
- Burr provides orchestration and execution visibility; developers supply model calls, API querying and data management.
- Pluggable persisters support saving and loading application state.
- The optional CLI in the documented extras requires Python 3.10 or later, while core library usage supports Python 3.9.
- The project describes execution on infrastructure managed by the developer.
Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.
Install Burr and view its telemetry UI
Install the documented package extra and start Burr’s UI with its sample data and demo.
Before you start
- Python 3.9 or later for core library usage; Python 3.10 or later for the optional CLI in the documented extras.
- A Python environment with package installation available.
- Set OPENAI_API_KEY only if you want to chat with the demo chatbot.
Steps
Install with the command below.
pip install "apache-burr[start]"Run
burrto start the telemetry UI.Open the Demos sidebar and select
chatbotto inspect the demo; provide OPENAI_API_KEY to chat with it.
Expected outcome
The telemetry UI opens with default data and a demo chatbot that can be inspected; chatting with the demo requires the documented API key.
Guide limitations
- The API key requirement applies to chatting with the demo, not viewing the UI.
- The optional CLI in the documented extras requires Python 3.10 or later; core library usage supports Python 3.9.
Run the bundled hello-world counter example
Run the repository’s counter example and view its execution trace in Burr’s UI.
Before you start
- Git and Python.
- Burr installed in the environment.
- The Burr telemetry UI started with
burr.
Steps
Clone
https://github.com/apache/burr.Change into
burr/examples/hello-world-counter.Run the command below.
python application.pyFind the counter execution trace in the telemetry UI.
Expected outcome
The README describes the counter running in the terminal with its trace tracked in the UI.
Guide limitations
- This guide follows the repository’s documented example; it does not specify additional environment or dependency setup.
Sources & review
Source record details
- Source availability
- Active
- Sources checked
- 13 October 2026
- Original website title
- apache/burr · Official homepage title · retained in its source language