Seeking Advice on Implementing Vector Search for Generative AI in MongoDB

Prisha_Singh · 2024-04-27T17:50:18.305Z

Hi everyone,

I’m currently working on a project involving generative AI and vector search, and I’m seeking some guidance on implementing this within MongoDB.

Specifically, I’m interested in exploring how to efficiently store and query large collections of embeddings generated by language models (LLMs) such as GPT-3 or BERT. My goal is to build a system that can perform similarity searches on these embeddings to retrieve relevant data points.

I’ve been reading up on MongoDB’s capabilities for handling vector data, but I’m still unsure about the best practices for modeling the data and optimizing the queries for performance.
I also check this : https://www.mongodb.com/community/forums/t/how-to-implement-filters-in-mongodb-atlas-vector-search-using-langchain generative ai

If anyone has experience or insights into implementing vector search for LLMs in MongoDB, I would greatly appreciate any advice or pointers you could provide.

Thank you in advance for your help!

Kushagra_Kesav · 2024-04-28T10:20:20.971Z

Hi @Prisha_Singh,

Welcome to the MongoDB Community forums

Prisha_Singh:

I’m interested in exploring how to efficiently store and query large collections of embeddings generated by language models (LLMs) such as GPT-3 or BERT.

I’ll recommend you to go through the tutorial by @Anaiya_Raisinghani on our Developer Center: How to Model Your Documents for Vector Search | MongoDB

It will give you an understanding of how to properly model your documents when using Atlas vector search to make your query efficient.

Additionally, you can also refer to our Developer Center for more content on the vector search: AI: MongoDB.

In case of any further questions feel free to reach out.

Best regards,
Kushagra