Skip to main content

Modules

Introduction​

LangChain is a framework for developing applications powered by large language models (LLMs). It provides components and tools to build chains - configurable sequences of calls to language models, data sources, and other services.

The key benefits of LangChain are:

  • Modularity: Components are designed to be modular and composable
  • Flexibility: Chains are customizable for different use cases
  • Accessibility: Standard interfaces make it easy to integrate LLMs

Get started by following the installation and quickstart guides.

Core Concepts​

LangChain is organized around several key concepts:

Modules​

Modules provide interfaces and integrations for core functionality like:

  • Model I/O: Interface with LLMs via prompts and parsers
  • Data Connection: Connect to external data sources
  • Chains: Construct sequences of calls to models, data, and services
  • Agents: Enable chains to operate autonomously based on directives
  • Memory: Persist state across chain executions
  • Callbacks: Log and stream chain steps
  • Evaluation: Assess chain performance

Chains​

Chains sequence calls to models, data sources, and services to accomplish a task. For example, a chain could:

  1. Query a database
  2. Pass results to a LLM
  3. Parse model outputs
  4. Return final response

Chains make it easy to combine different capabilities in a customizable pipeline.

Directives​

Directives provide high-level instructions that guide a chain. Agents execute chains based on directives.

For example, a directive could be: "Answer the question based on the provided context". The agent would then construct and run a chain to accomplish this using the available tools.

Examples​

LangChain supports common use cases like:

See the use cases section for details and code examples.

The gallery contains many sample projects built with LangChain.

Get Involved​