BeeAI Framework (formerly Bee Agent Framework) is an open-source framework from IBM Research, now hosted under Linux Foundation open governance, for building production-ready multi-agent systems. It supports both Python and TypeScript with full feature parity, is model-agnostic (Ollama, Watsonx.ai, OpenAI, Groq, and 10+ providers), and includes built-in memory management, MCP tool integration, caching, and resource optimization. The framework emphasizes interoperability, allowing agents from different implementations to collaborate on the same workflow. Key features: - Python and TypeScript support with complete feature parity - Model-agnostic: 10+ LLM providers including Watsonx.ai, OpenAI, Ollama, Groq - Multi-agent orchestration with cross-implementation agent collaboration - Four memory strategy types optimized for different workload patterns - MCP (Model Context Protocol) tool integration support - Production features: caching, resource management, built-in observability
Free and open source (Apache 2.0). Hosted under Linux Foundation.