openai-cookbook/examples/vector_databases/weaviate/README.md

Weaviate <> OpenAI

​Weaviate is an open-source vector search engine (docs - Github) that can store and search through OpenAI embeddings and data objects. The database allows you to do similarity search, hybrid search (the combining of multiple search techniques, such as keyword-based and vector search), and generative search (like Q&A). Weaviate also supports a wide variety of OpenAI-based modules (e.g., text2vec-openai, qna-openai), allowing you to vectorize and query data fast and efficiently.

You can run Weaviate (including the OpenAI modules if desired) in three ways:

1. Open source inside a Docker-container (example)
2. Using the Weaviate Cloud Service (get started)
3. In a Kubernetes cluster (learn more)

Examples

This folder contains a variety of Weaviate and OpenAI examples.

Name	Description	lanugage	Google Colab
Getting Started with Weaviate and OpenAI	A simple getting started for semantic vector search using the OpenAI vectorization module in Weaviate (text2vec-openai)	Python Notebook	link
Hybrid Search with Weaviate and OpenAI	A simple getting started for hybrid search using the OpenAI vectorization module in Weaviate (text2vec-openai)	Python Notebook	link
Question Answering with Weaviate and OpenAI	A simple getting started for question answering (Q&A) using the OpenAI Q&A module in Weaviate (qna-openai)	Python Notebook	link
Docker-compose example	A Docker-compose file with all OpenAI modules enabled	Docker