LangGraph

LangChain 的显式图编排框架，隐式循环的对比范式

实

ENTITY · LANGGRAPH · LANGCHAIN · EXPLICIT GRAPH · COUNTERPART TO THE IMPLICIT LOOP

LangGraph

LangChain’s low-level agent-orchestration framework — defines a directed graph of nodes and edges via StateGraph

LangGraph is the main contrast paradigm to the implicit-loop architecture (Claude Agent SDK, Codex), representing one of the two poles of agent orchestration. Explicit-graph architecture: the flow is fixed at compile time, StateGraph defines nodes and edges of a directed graph, giving high predictability and easy debugging. Adopted by enterprises including Klarna, Uber, and J.P. Morgan, with durable execution, human-in-the-loop, and memory management as headline features.

Two orchestration paradigms compared

LangGraph (explicit graph)Flow fixed at compile timeHigh predictability, easy to debugLower flexibility, relies on pre-set paths

Implicit loop (Claude/Codex)Model decides autonomously at runtimeHigh flexibility, no pre-set paths neededLower predictability, relies on model capability

LangGraph signature features

Durable Execution

Execution state is persistable — tasks can resume from the breakpoint after interruption, fitting long-running workloads

Human-in-the-loop

Built-in support for human intervention — conditional nodes can pause and wait for a human decision

Enterprise adoption

Klarna / Uber / J.P. Morgan — the auditability of explicit graphs suits enterprise compliance needs

→ Implicit Loop Architecture · Claude Agent SDK · CodexLangGraph Documentation

LangGraph

LangChain 提供的低层 agent 编排框架，采用显式图架构——用 StateGraph 定义节点和边的有向图来编排 agent 行为。

与本 wiki 的关联

LangGraph 是隐式循环架构（Claude Agent SDK、Codex）的主要对比范式。两者代表了 agent 编排的两极：

LangGraph：流程在编译时确定，可预测性高，调试容易
隐式循环：模型在运行时自主决策，灵活性高，依赖模型能力

被 Klarna、Uber、J.P. Morgan 等企业采用，主打 durable execution、human-in-the-loop、内存管理。

References

sources/langgraph-documentation.md

LangGraph

LangGraph

与本 wiki 的关联

相关实体

References