LangGraph

Graph-based framework for building stateful, controllable agent workflows.

Overview

LangGraph is a library for building stateful, multi-actor applications with language models. Developers define workflow state, nodes, and graph edges to control agent execution. Checkpointing supports persistence, while interrupts enable human review and intervention. LangGraph is available for Python and JavaScript and can be used to build single-agent or multi-agent systems.

Key features

Explicit graph-based control flow
State schemas and partial updates
Checkpointing and persistence
Human-in-the-loop interrupts
Multi-actor workflow support
Python and JavaScript packages

Use cases

Show all 8Show fewer

Alternatives

CofounderAI agent orchestration platform for running company operations across departmentsai agent
AgentForgeOpen-source low-code Python framework for agents and YAML cognitive architectures.ai agent frameworkFree
AgnoSelf-hosted Python agent platform with AgentOS runtime and browser control plane.ai agent frameworkFreemium
AutoGenMicrosoft open-source framework for multi-agent chat apps and event-driven agent systems.ai agent frameworkFree
BondAIOpen-source Python framework and CLI for single- and multi-agent systems with built-in tools.ai agent frameworkFree
CrewAIOpen-source framework and platform for multi-agent crews and event-driven flows.ai agent frameworkFreemium

Last updated 2026-09-16