The langchain-teradata package is a powerful Python integration designed to seamlessly connect LangChain applications with Teradata Vantage. It enables developers to leverage Teradata’s high-performance Enterprise Vector Store within the LangChain framework, streamlining the development of Retrieval-Augmented Generation (RAG) and other LLM-powered workflows.
This document provides detailed description and complete usage information for all the functions in the Teradata® Package for LangChain, langchain-teradata.
What You Can Do
-
You can use the langchain-teradata package to make use of the following vector store management capabilities:
- Vector Store creation from documents, raw text, structured datasets, or pregenerated embeddings
- Document, Dataset, and Embedding Data ingestion and management
- Retriever generation for RAG workflows
- Incremental updates to existing vector stores
- Language Translation
- Deletion of documents, datasets, and embeddings
-
Use these functions to build scalable, LLM-powered applications using LangChain's standardized interfaces while leveraging Teradata's high-performance Enterprise Vector Store.
-
Enterprise Vector Store is specifically designed for the efficient storage, indexing, and searching of high-dimensional vector embeddings, making it ideal for Retrieval-Augmented Generation (RAG), semantic search, and other advanced AI use cases.
Get Started
-
Verify the following requirements to use Teradata Package for LangChain:
- Teradata Vantage with database release 20.00.29.XX or later
- Vector Store (Data Insights) is enabled
[See _(Install, Upgrade, and Unistall] (https://docs.teradata.com/access/sources/dita/topic?dita:topicPath=zfw1738940533087.dita)