LangChain is the most popular open-source framework for building LLM-powered applications. Supports agents, RAG, and chains. Stars 100k+.
CrewAI is a lightweight multi-agent orchestration framework. Define agent roles, assign tasks, and run collaborative AI teams. Stars 25k+.
Microsoft's open-source multi-agent conversation framework. Enables building LLM applications with multiple agents that converse and collaborate.
Open-source LLM application development platform. Visual workflow builder for creating AI agents, chatbots, and RAG applications.
AI bot building platform by ByteDance. Drag-and-drop agent creation with built-in plugins and knowledge base.
Framework for building stateful multi-agent applications from the LangChain team. Supports cyclic graphs.
Microsoft's lightweight AI orchestration SDK for .NET and Python. Integrates LLMs with traditional programming for agentic workflows.
Agent framework by the Pydantic team. Type-safe, structured output guaranteed — built for production-grade AI agents.
Multi-modal agent framework (formerly Phidata). Build agents that can see, hear, and interact with any data source.
Autonomous AI agent framework backed by a16z. Designed for creating personality-driven agents with memory and tool use.
Based on hedirbase's curated directory, the most notable AI agent frameworks in 2026 include LangChain, CrewAI, AutoGen, Dify, Coze, LangGraph, Semantic Kernel, Pydantic AI, Agno, Eliza. Each framework takes a different approach: orchestration layers, multi-agent crews, or visual builders.
The answer depends on your background: developers comfortable with Python often start with LangChain or CrewAI, while non-programmers prefer visual builders. All frameworks on this list have active open-source communities.
hedirbase verifies the links and repo URLs of listed frameworks on a regular basis and records a last-verified date on every entry page. This ranking reflects the current directory data; new frameworks can be suggested through the Submit page at any time.