LangGraph is a library for building stateful, controllable agents as graphs, giving you fine-grained control over loops, branching and persistence.
| Category | AI agent framework |
| Type | Agent orchestration (graphs) |
| License | MIT |
| Runs locally | Cloud-optional |
| Built with | Python / JS |
| Skill level | Advanced |
| Best for | developers needing controllable agent workflows |
Other open-source ai agent framework tools worth comparing:
CrewAIOrchestrate role-playing agent crews
PaperclipRun a company of AI agents from one dashboard
AutoGenMicrosoft's conversational agent framework
MetaGPTA software company of LLM agents
AutoGPTThe original autonomous GPT agent
OpenHandsAI developer that writes and runs code
SuperAGIDev-first autonomous agent platform
BabyAGIMinimal task-loop autonomous agent
AgnoFast, lightweight multi-modal agents
LettaStateful agents with long-term memory
smolagentsAgents in a few lines of code
Pydantic AIType-safe agents for production
SWE-agentAgent that fixes GitHub issues
Agent ZeroA general agent that grows with your prompts
CAMELThe research framework for agent societies
ElizaAgents with personality for Discord, X and more
Google ADKGoogle's official Agent Development Kit
Mem0Long-term memory for AI agents
Composio250+ ready-made tools for your agents
AG2The community continuation of AutoGen
OpenAI SwarmMinimal multi-agent handoffs
PhidataAgents with memory, knowledge and tools
JulepStateful agents as a serviceLangGraph is free and open-source (MIT license), so you can use, self-host and modify it at no cost.
LangGraph can be used locally or in the cloud depending on your setup.
Popular open-source alternatives include CrewAI, Paperclip, AutoGen. See the comparisons above to choose.
Browse the full directory of open-source AI tools, models and projects — updated daily.
Browse all tools →