Components

LangChain provides standard, extendable interfaces and external integrations for the following main components:

Model I/O

Formatting and managing language model input and output

Formatting for LLM inputs that guide generation

Interfaces for language models that use chat messages as inputs and returns chat messages as outputs (as opposed to using plain text).

Interfaces for language models that use plain text as input and output

Interface with application-specific data for e.g. RAG

Load data from a source as Documents for later processing

Transform source documents to better suit your application

Create vector representations of a piece of text, allowing for natural language search

Interfaces for specialized databases that can search over unstructured data with natural language

More generic interfaces that return documents given an unstructured query

Higher-level components that combine other arbitrary systems and/or or LangChain primitives together

Interfaces that allow an LLM to interact with external systems

Constructs that choose which tools to use given high-level directives

Building block-style compositions of other runnables

Persist application state between runs of a chain

Log and stream intermediate steps of any chain