Langchain
Transform your LangChain applications with Mixedbread's AI-native search platform. Seamlessly integrate multimodal semantic understanding, state-of-the-art embeddings, and intelligent reranking into your LangChain chains and LCEL workflows.
Quick Start
- Install the package:
- Save your API key to an environment variable:
Components
This integration provides four LangChain-compatible components for building advanced RAG chains:
Embeddings
: Transform text into semantic vectors using state-of-the-art models, compatible with all LangChain vector stores and retrieval chainsReranker
: Enhance retrieval quality by semantically reordering documents in your RAG pipelines, seamlessly integrating with LangChain's document transformersDocumentLoader
: Parse multimodal documents (PDF, PPTX, HTML) with layout-aware intelligence, returning structured LangChain Document objectsVectorStoreRetriever
: Query Mixedbread's managed vector stores directly within your LangChain retrieval chains and LCEL expressions
Embeddings
Generate text embeddings for queries and documents:
Reranker
Rerank documents based on query relevance:
Document Loader
Load and parse documents from various formats:
Vector Store Retriever
Retrieve documents from Mixedbread vector stores:
Chain Examples
RAG Chain with vector store retriever
Delete File
Delete a specific file by its ID.
Haystack
Enhance your Haystack NLP pipelines with Mixedbread's semantic intelligence platform. This guide covers modular component integration, enterprise-ready workflows, and production-scale document processing using advanced multimodal AI models.
Last updated: August 18, 2025