Best Python AI Agent Frameworks

Open-source framework with a standard interface over many model providers and a prebuilt agent harness (create_agent) built on LangGraph, available for Python and JavaScript/TypeScript.

FreeLibraryOpen source★ 148,264

Open-source Python framework for orchestrating role-based multi-agent 'crews' and event-driven Flows, with an optional hosted platform for building and running them.

Free tier availableLibraryOpen source★ 59,981

Open-source Python framework for building LLM agents and retrieval (RAG) pipelines over your own data, from the company behind the LlamaParse document-parsing platform.

FreeLibraryOpen source★ 52,617

Low-level open-source orchestration framework and runtime for long-running, stateful agents modeled as graphs, with durable execution, streaming, and human-in-the-loop support.

Free tier, paid from $39/moLibraryOpen source★ 43,714

Google's open-source, code-first toolkit for building, evaluating, and deploying single- and multi-agent systems, optimized for Gemini and Google Cloud but usable with other models.

FreeLibraryOpen source★ 21,961

Type-safe Python agent framework from the Pydantic team, with validated structured outputs, many model providers selectable by string, and optional durable execution and evals.

FreeLibraryOpen source★ 20,787

AWS's open-source, model-driven SDK for building AI agents in Python and TypeScript, where the model plans and calls tools in a built-in agent loop; it now also ships a pre-assembled 'Strands harness' agent.

FreeLibraryOpen source★ 8,911

Microsoft's open-source framework for multi-agent conversational AI applications. It is no longer maintained, and Microsoft directs users to Microsoft Agent Framework.

FreeLibraryOpen source★ 61,522