Skip to content

Guides

The pages in this section provide a conceptual overview and how-tos for the following topics:

Agent development

  • Overview: Use prebuilt components to build an agent.
  • Run an agent: Run an agent by providing input, interpreting output, enabling streaming, and controlling execution limits.

LangGraph APIs

  • Graph API: Use the Graph API to define workflows using a graph paradigm.
  • Functional API: Use Functional API to build workflows using a functional paradigm without thinking about the graph structure.
  • Runtime: Pregel implements LangGraph's runtime, managing the execution of LangGraph applications.

Core capabilities

These capabilities are available in both LangGraph OSS and the LangGraph Platform.

  • Streaming: Stream outputs from a LangGraph graph.
  • Persistence: Persist the state of a LangGraph graph.
  • Durable execution: Save progress at key points in the graph execution.
  • Memory: Remember information about previous interactions.
  • Context: Pass outside data to a LangGraph graph to provide context for the graph execution.
  • Models: Integrate various LLMs into your LangGraph application.
  • Tools: Interface directly with external systems.
  • Human-in-the-loop: Pause a graph and wait for human input at any point in a workflow.
  • Time travel: Travel back in time to a specific point in the execution of a LangGraph graph.
  • Subgraphs: Build modular graphs.
  • Multi-agent: Break down a complex workflow into multiple agents.
  • MCP: Use MCP servers in a LangGraph graph.
  • Evaluation: Use LangSmith to evaluate your graph's performance.

Platform-only capabilities

These capabilities are only available in LangGraph Platform.