BeeAI Framework
IBM's open-source framework for production-grade multi-agent systems, hosted by the Linux Foundation.
About
BeeAI Framework is an open-source framework for building production-grade multi-agent systems, contributed by IBM Research to the Linux Foundation's LF AI & Data program. It offers complete feature parity between its Python and TypeScript implementations, with workflow composition, advanced memory strategies, provider-agnostic backends (OpenAI, Ollama, watsonx.ai and more), MCP tool integration, and full observability via OpenTelemetry.
BeeAI is also the home of the Agent Communication Protocol (ACP), IBM's open REST-based protocol for agent-to-agent interoperability, which in 2025 merged into the Linux Foundation's Agent-to-Agent (A2A) effort. A companion BeeAI Platform lets developers discover, run, and share agents across frameworks like CrewAI, LangGraph, and AutoGen.
What it can do
- Build multi-agent workflows in Python or TypeScript with full feature parity
- Provider-agnostic backend: OpenAI, Ollama, watsonx.ai and 10+ others
- Four memory strategies plus custom memory and OpenTelemetry tracing
- Discover and compose agents across frameworks via the ACP protocol
Setup
pip install beeai-framework (Python) or npm install beeai-framework (TypeScript); no signup required.
Prerequisites
Python 3.11+ or Node 18+, an API key for your chosen LLM provider.
Use cases
Works with
Trust score — 8/10
Scored 8/10 (verified tier, metrics-v1). Evidence: Verified via vendor site/docs/GitHub and reputable coverage, wave 73; Trending score 82/100 from researcher assessment.
Sources
Every fact above was checked against these sources. Where a claim comes from the provider and couldn't be independently verified, the text says so.