Haystack is an open source framework by deepset for building production-ready LLM applications, retrieval-augmented generative pipelines and state-of-the-art search systems that work intelligently over large document collections. It lets you quickly try out the latest models in natural language processing (NLP) while being flexible and easy to use.

This is the repository where we keep all the Haystack tutorials 📓 👇 These tutorials are also published to the Haystack Website.

To contribute to the tutorials, please check out our Contributing Guidelines.

Haystack 1.x	Haystack 2.0

Code	Colab	Code	Colab
Build Your First Question Answering System		Your First QA Pipeline with Retrieval-Augmentation
Fine Tune a Model on Your Data		Generating Structured Output with Loop-Based Auto-Correction
Build a Scalable Question Answering System		Serializing Pipelines
FAQ Style QA		Preprocessing Different File Types
Evaluation		Metadata Filtering
Better Retrieval via Embedding Retrieval		Classifying Documents & Queries by Language
[OUTDATED] RAG Generator		Creating a Hybrid Retrieval Pipeline
Preprocessing		Build an Extractive QA Pipeline
DPR Training		Model-Based Evaluation of RAG Pipelines
[OUTDATED] Knowledge Graph		Building Pipelines with Conditional Routing
Pipelines		Simplifying Pipeline Inputs with Multiplexer
[OUTDATED] Seq2SeqGenerator		Embedding Metadata for Improved Retrieval
Question Generation		Building a Chat Application with Function Calling
Query Classifier
Table QA
Document Classifier at Index Time
Make Your QA Pipelines Talk!
Generative Pseudo Labeling
Text-to-Image search
Using Haystack with REST API
Customizing PromptNode
Generative QA Pipeline with Retrieval-Augmentation
Answering Complex Questions with Agents
Building a Conversational Chat App
Customizing Agent to Chat with Your Documents
Creating a Hybrid Retrieval Pipeline