Semantic Search with OpenAI and Upstash Vector

Welcome!

This tutorial will guide you through creating similarity search application using Upstash Vector and OpenAI Embeddings API. We will create the embeddings using the pre-prepared dataset that contains 100 questions. Then we’ll upsert these vectors to Upstash, both to store the vectors, and to use basic vector operations such as query, fetch etc. Here is the general overview of the project.

You can use this example as a template to build more advanced projects such as docs search, QA bot, similarity search over a content etc.

Outline:

1- Create Index and install dependencies

2- Generate embeddings

3- Insert vectors to index

4- Query the index

5- Outro

You can find the full tutorial and code in the notebook here.