Repolode
Submit
i-am-bee

i-am-bee/beeai-framework

A Python and TypeScript framework for building multi-agent AI systems with LLM backends, tools, memory, RAG and workflows.

View on GitHubWebsite
Stars
3,452
Forks
543
Open issues
11
Last commit
5 days ago

What is it?

BeeAI Framework is a toolkit for building AI agents that can reason, call tools, and collaborate with other agents. It is available as a Python library and as a TypeScript library, with a unified backend interface for connecting to different LLM providers.

The framework includes agents with rule-based behavior controls, built-in tools such as web search and weather lookup, custom tool support, and RAG components built on vector stores and document processing. It also provides prompt templates, conversation memory strategies, caching, state serialization, and workflows for orchestrating multi-agent execution. Agents can be served over protocols such as A2A and MCP.

Developers working on multi-agent applications can use it to route questions between specialist agents through handoff tools, or to build agents whose tool use must follow defined requirements. Event, logging, and error-handling hooks are provided for observing agent behavior.

Key features

  • 01Requirement Agent that enforces rules on tool use across different LLMs.
  • 02Unified backend interface for multiple LLM providers in Python and TypeScript.
  • 03Built-in tools for web search, weather and code execution, plus custom tools.
  • 04RAG components with vector stores and document processing.
  • 05Workflows for orchestrating multi-agent execution flows.
  • 06Serving agents over A2A and MCP protocols.

Use cases

  • Building a router agent that hands questions off to specialist agents.
  • Creating LLM agents with defined tool-use rules for predictable behavior.
  • Adding retrieval-augmented generation to an application with document ingestion.
  • Running multi-agent workflows that persist state across sessions.

Quick start

  1. 1

    Install the Python library

    pip install beeai-framework
  2. 2

    Install the TypeScript library (alternative)

    npm install beeai-framework
  3. 3

    Install ollama and download the granite4.1:8b model, then run your project file

    python [project_name].py

Things to know

  • The legal notice says IBM will not maintain this code going forward and makes no quality or security claims.
  • The multi-agent example needs a local ollama install with the granite4.1:8b model, or another configured LLM provider.
  • The Requirement Agent is experimental, and the Python library was launched as an alpha.

In collections

Multi-agent orchestration frameworksOpen-source AI agent frameworks & apps

Topics

#agents#multi-agent#llm#python#typescript#rag#workflows#ai#ai-agent#beeai

FAQ

What is beeai-framework?
A Python and TypeScript framework for building multi-agent AI systems with LLM backends, tools, memory, RAG and workflows. BeeAI Framework is a toolkit for building AI agents that can reason, call tools, and collaborate with other agents. It is available as a Python library and as a TypeScript library, with a unified backend interface for connecting to different LLM providers.
Is beeai-framework free to use? What license does it have?
beeai-framework is open source under the Apache-2.0 license. Check the license terms before using it in commercial products.
Is beeai-framework still maintained?
Actively maintained. The last commit to beeai-framework was on Nov 3, 2026.
What can I use beeai-framework for?
Building a router agent that hands questions off to specialist agents. Creating LLM agents with defined tool-use rules for predictable behavior. Adding retrieval-augmented generation to an application with document ingestion. Running multi-agent workflows that persist state across sessions.
How do I install beeai-framework?
Install the Python library: pip install beeai-framework Install the TypeScript library (alternative): npm install beeai-framework Install ollama and download the granite4.1:8b model, then run your project file: python [project_name].py
Where can I get beeai-framework?
The source code and installation instructions are on GitHub: https://github.com/i-am-bee/beeai-framework

Summary generated by AI from the project's README. GitHub stats refresh daily.

Related projects