Burr Framework

Open-source Python framework for building stateful, decision-making applications like agents and chatbots.

4.3 (4)

Beoordeeld door Daniel Nikulshyn·Bijgewerkt mei 2026

Observability Python Framework Open Source AI Agents State Management LLM Developer Tools

Overzicht

Burr Framework is a Python library for building applications that need to make decisions over time, such as chatbots, AI agents, simulations, and workflow engines. It models programs as state machines, letting developers define actions and transitions that operate on a shared state object, making complex control flow easier to reason about. The framework includes built-in observability tools, a local UI for inspecting runs, and support for persistence so applications can pause, resume, and be debugged step by step. Because Burr is unopinionated about which LLMs or libraries you use, it integrates with most of the popular Python AI stack. It is well suited for teams that want explicit control over agent logic rather than relying on black-box orchestration, and for production systems where traceability and testability matter.

Belangrijkste functies

State machine abstraction with actions and transitions
Local telemetry UI for inspecting executions
State persistence and resumability
Streaming and async action support
Integrations with common LLM and ML tools
Hooks for logging, monitoring, and testing

Use cases

Build stateful chatbots with traceable logic

Model conversational flows as explicit state machines with actions and transitions, making it easier to reason about chatbot behavior and debug runs via the local telemetry UI.

Develop decision-making AI agents

Create AI agents that manage shared state across steps, with support for streaming, async actions, and integration with any LLM library in the Python ecosystem.

Run resumable workflow engines

Use state persistence to pause, resume, and step-debug long-running workflows or simulations, enabling reliable recovery and inspection of complex control flow.

Instrument AI apps for monitoring and testing

Leverage built-in hooks for logging, monitoring, and tracing to observe production AI applications and validate behavior through reproducible, inspectable runs.

Pluspunten & minpunten

Pluspunten

Explicit state-machine model makes logic easy to follow
Built-in tracing UI for debugging runs
Framework-agnostic—works with any LLM or library
Supports persistence, streaming, and async
Open source and lightweight

Minpunten

Requires Python and some learning of its abstractions
Less plug-and-play than higher-level agent frameworks
Smaller community than larger competitors

Reviews

4.3

Gemiddelde van 4 beoordelingen.

Priya Nair

Does the job

Pretty happy overall. Local telemetry UI for inspecting executions just works and built-in tracing UI for debugging runs. Less plug-and-play than higher-level agent frameworks can be annoying, but no dealbreakers — I'd recommend it to a friend without hesitating.

Ethan Brooks

Compared a few options

Evaluated this against two competitors. Where it wins: local telemetry UI for inspecting executions and explicit state-machine model makes logic easy to follow. Where it lags: requires Python and some learning of its abstractions. On balance the feature set — especially local telemetry UI for inspecting executions — justifies the 5 stars for our use case.

Tariq Aziz

Use it every day

Honestly didn't expect to like it this much. State persistence and resumability is exactly what I needed, and open source and lightweight. I do wish smaller community than larger competitors, but I reach for it almost every day now and it just clicks.

Diego Fernández

Solid for our team

We rolled this out across the team last quarter and built-in tracing UI for debugging runs. State persistence and resumability fits neatly into how we already work, and integrations with common LLM and ML tools removed a step we used to do by hand. Smaller community than larger competitors, which is the main caveat, but it has held up under daily use.