[{"data":1,"prerenderedAt":12},["ShallowReactive",2],{"stack-tr-en-langgraph":3},{"tagline":4,"description":5,"strengths":6,"useCase":11},"LangChain's runtime for stateful AI agents, orchestrated as graphs.","LangGraph is LangChain's library for orchestrating agents (and multi-agent systems) as graphs: nodes, conditional edges, loops and shared state. Checkpointing makes every run persistent and resumable, interrupts enable human validation, and streaming exposes progress in real time. It's our choice when an agent outgrows the linear chain: decisions, loops and human oversight in production.",[7,8,9,10],"Explicit graphs: nodes, conditional edges, loops, multi-agent.","Persistence and checkpointing: resumable runs, durable memory.","Human-in-the-loop: interrupts to validate or correct a step.","Native streaming of tokens and steps for responsive interfaces.","Supervised autonomous agents, multi-agent workflows, long-running business assistants.",1792675978998]