Integrations

Amazon Bedrock

Use Models from AI21 Labs, Anthropic, Cohere, Meta, and Amazon via Amazon Bedrock with Haystack

Model Provider

Maintained by deepset

Amazon Sagemaker

Use Models from Huggingface, Anthropic, AI21 Labs, Cohere, Meta, and Amazon via Amazon Sagemaker with Haystack

Model Provider

Maintained by deepset

Anthropic

Use Anthropic Models with Haystack

Model Provider

Maintained by deepset

Apify

Extract data from the web and automate web tasks using Apify-Haystack integration.

Data Ingestion

Arize Phoenix

Trace your Haystack pipelines with Arize Phoenix

Monitoring Tool

Arize AI

Trace and Monitor your Haystack pipelines with Arize AI

Monitoring Tool

AssemblyAI

Use AssemblyAI transcription, summarization and speaker diarization models with Haystack

Model Provider

AstraDB

A Document Store for storing and retrieval from AstraDB.

Document Store

Maintained by deepset

Azure AI Search

Use Azure AI Search with Haystack

Document Store

Maintained by deepset

Azure CosmosDB

Use Azure CosmosDB with Haystack

Document Store

Maintained by deepset

Azure

Use OpenAI models deployed through Azure services with Haystack

Model Provider

Maintained by deepset

Burr

Build an agent from Haystack components

Framework

Cerebras

Use LLMs served by Cerebras API

Model Provider

Maintained by deepset

Chroma

A Document Store for storing and retrieval from Chroma

Document Store

Maintained by deepset

Cohere

Use Cohere models with Haystack

Model Provider

Maintained by deepset

Context AI

A component to log conversations for analytics by Context.ai

Monitoring Tool

Couchbase

Use the Couchbase database with Haystack

Document Store

DeepEval

Use the DeepEval evaluation framework to calculate model-based metrics

Evaluation Framework

Maintained by deepset

DeepL

Use DeepL translation services with Haystack

Custom Component

Docling

Use Docling to locally parse and chunk PDF, DOCX, and other document types in Haystack

Data Ingestion

DuckDuckGo

Uses DuckDuckGo API for web searches

Data Ingestion

Elasticsearch

Use an Elasticsearch database with Haystack

Document Store

Maintained by deepset

Elevenlabs

ElevenLabs Text-to-Speech components for Haystack.

Model Provider

FastEmbed

Use the FastEmbed embedding models

Model Provider

Maintained by deepset

fastRAG

fastRAG is a research framework for efficient and optimized retrieval augmented generative pipelines

Custom Component

Flow Judge

Evaluate Haystack pipelines using Flow Judge

Evaluation Framework

Google AI

Use Google AI Models with Haystack

Model Provider

Maintained by deepset

Google Vertex AI

Use Google Vertex AI Models with Haystack

Model Provider

Maintained by deepset

Groq

Use open Language Models served by Groq

Model Provider

Maintained by deepset

Hugging Face

Use Models on Hugging Face with Haystack

Model Provider

Maintained by deepset

INSTRUCTOR Embedders

A component for computing embeddings using INSTRUCTOR embedding models.

Model Provider

Jina AI

Use the latest Jina AI embedding models

Model Provider

Maintained by deepset

LanceDB Haystack

A DocumentStore backed by LanceDB

Document Store

langfuse

Monitor and trace your Haystack requests.

Monitoring Tool

Maintained by deepset

Llama.cpp

Use Llama.cpp models with Haystack.

Model Provider

llamafile

Run LLMs locally with llamafile

Model Provider

Maintained by deepset

LM Format Enforcer

Use the LM Format Enforcer to enforce JSON Schema / Regex output of your Local Models.

Model Provider

Marqo

A Document Store for storing and retrieval from Marqo

Document Store

Mastodon Fetcher

A custom component to fetch a mastodon usernames latest posts

Data Ingestion

Model Context Protocol - MCP

Haystack Tool Integration with the MCP

Tool Integration

Maintained by deepset

Milvus

Use the Milvus vector database with Haystack

Document Store

Mistral

Use the Mistral API for embedding and text generation models.

Model Provider

Maintained by deepset

mixedbread ai

Use mixedbread's models as well as top open-source models in seconds

Model Provider

MongoDB

Use a MongoDB Atlas database with Haystack

Document Store

Maintained by deepset

MonsterAPI

Use open Language Models served by MonsterAPI

Model Provider

Needle

Use Needle document store and retriever in Haystack.

Document Store

Neo4j

Use the Neo4j database with Haystack

Document Store

Notion Extractor

A component to extract pages from Notion to Haystack Documents. Useful for indexing Pipelines.

Data Ingestion

NVIDIA

Use NVIDIA models with Haystack.

Model Provider

Maintained by deepset

Ollama

Use Ollama models with Haystack. Ollama allows you to get up and running with large language models, locally.

Model Provider

Maintained by deepset

OpenAI

Use OpenAI Models with Haystack

Model Provider

Maintained by deepset

OpenSearch

A Document Store for storing and retrieval from OpenSearch

Document Store

Maintained by deepset

Opik

Trace and evaluate your Haystack pipelines with Opik

Monitoring Tool

Optimum

High-performance inference using Hugging Face Optimum

Model Provider

Maintained by deepset

pgvector

A Document Store for storing and retrieval from pgvector

Document Store

Maintained by deepset

Pinecone

Use a Pinecone database with Haystack

Document Store

Maintained by deepset

Qdrant

Use the Qdrant vector database with Haystack

Document Store

Maintained by deepset

Ragas

Use the Ragas evaluation framework to calculate model-based metrics

Evaluation Framework

Ray

Run and scale Haystack Pipelines with Ray in distributed manner

Distributed Computing

Sambanova

Use open language models served by Sambanova

Model Provider

Snowflake

A Snowflake integration that allows table retrieval from a Snowflake database.

Data Ingestion

Maintained by deepset

STACKIT

Use the STACKIT API for text generation models.

Model Provider

Maintained by deepset

Titan Takeoff Inference Server

Use Titan Takeoff to run local open-source LLMs with Haystack. Titan Takeoff allows you to run the latest models from Meta, Mistral and Alphabet directly in your laptop.

Model Provider

Traceloop

Evaluate and monitor the quality of your LLM apps and agents

Monitoring Tool

Trafilatura

Efficiently gather text and metadata on the Web for LLM and RAG

Data Ingestion

Unstructured File Converter

Component to easily convert files and directories into Documents using the Unstructured API

Data Ingestion

Maintained by deepset

UpTrain

Use the UpTrain evaluation framework to calculate model-based metrics

Evaluation Framework

Maintained by deepset

vLLM Invocation Layer

Use the vLLM inference engine with Haystack

Model Provider

Voyage AI

Use text embeddings and rerankers from Voyage AI

Model Provider

Weaviate

Use a Weaviate database with Haystack

Document Store

Maintained by deepset

Weights & Biases Weave Tracer

Send Haystack traces to Weights & Biases for monitoring and visualization

Monitoring Tool

Maintained by deepset