LangGraph
Partial infoBuild stateful, controllable agents as graphs.
LangGraph models agent workflows as explicit graphs with persistent state, enabling controllable multi-step and multi-agent systems, human-in-the-loop, and streaming. It is open-source (MIT) with an optional managed platform for deployment.
License
Open source (MIT)
Deployment
Self-hosted, Managed
Pricing
Open-source; LangGraph Platform (managed deploy) is paid.
SDKs / languages
Python, JavaScript
Founded
2024
Strengths
- Explicit, controllable graph model
- Durable state + human-in-the-loop
- Strong for complex workflows
Limitations
- More concepts to learn
- Graph model can feel heavy for simple agents
Alternatives
AutoGen
Microsoft's framework for multi-agent conversations.
CrewAI
Role-based multi-agent orchestration in Python.
Dify
Low-code platform to build LLM apps and agents.
Google Agent Development Kit
Google's open-source kit for building and deploying agents.
Letta (MemGPT)
Stateful agents with long-term memory (MemGPT).
LlamaIndex
Data framework for RAG and agents over your data.
Facts verified 19 July 2026. Neutral summary, not an endorsement; verify current details with the vendor.