Haystack

Haystack

Haystack by deepset is a production-grade framework for building end-to-end NLP pipelines. It excels at RAG (retrieval-augmented generation), document search, and question answering, with first-class support for pipelines-as-code and a rich component ecosystem.

Key Features

• Pipeline architecture — DAG-based pipelines connecting components: retrievers, readers, generators, routers
• RAG out of the box — combine document stores, embedding retrievers, and generative models in minutes
• 40+ document stores — Elasticsearch, OpenSearch, Qdrant, Weaviate, Chroma, FAISS, and more
• Multi-modal — text, tables, and images
• Evaluation — built-in pipeline evaluation with RAGAS and custom metrics
• REST API — hayhooks turns any pipeline into a deployable HTTP service

Quick Start

pip install haystack-ai

from haystack import Pipeline
from haystack.components.generators import OpenAIGenerator
from haystack.components.builders import PromptBuilder

template = "Answer: {{question}}"
pipe = Pipeline()
pipe.add_component("prompt", PromptBuilder(template=template))
pipe.add_component("llm", OpenAIGenerator(model="gpt-4o-mini"))
pipe.connect("prompt", "llm")
result = pipe.run({"prompt": {"question": "Who invented the telephone?"}})
print(result["llm"]["replies"][0])

Install via ai-supply

npx ai-supply add haystack-nlp-pipeline

Curated mirror of the open-source Haystack project (Apache-2.0). Install upstream from the repository.