DocumentationAPI ReferenceTutorialsGitHub Code ExamplesDiscord Community

Haystack Concepts Overview

Haystack provides all the tools you need to build a custom NLP application that works for you. This includes everything from prototyping, through annotating data, to deployment. This page lists all the concepts Haystack operates on.

AgentUses a large language model to reason and answer complex queries that require multiple steps to find the correct answer.
Documents, Answers, and LabelsThe core data structures of Haystack.
Document StoresThese store your Document corpus and connect to various backend databases.
PipelinesPipelines allow for easy configuration of Haystack's NLP components.
Ready-Made PipelinesTemplate Pipelines that cover the most common use cases.
REST APIThe REST API allows for quick deployment of pipelines as a service.
Annotation ToolA tool for managing teams of annotators and facilitating the annotation of text for question answering and document search.
Pseudo Label GeneratorUse Pseudo Label Generator to create training data for dense Retrievers without human annotation.