Unlocking the power of Open AI and pgvector with Azure PostgreSQL and Azure Cosmos DB for PostgreSQL
Published May 24 2023 07:00 AM 12.6K Views
Microsoft

Azure Database for PostgreSQL Flexible Server and Azure Cosmos DB for PostgreSQL have now introduced support for the pgvector extension. With the pgvector extension, customers can now store embeddings in PostgreSQL databases which are vectors created by generative AI models that represent the semantic meaning of textual data allowing efficient similarity searches. This ensures that users can leverage the benefits of simplified model integration across different deployment options in Azure.

 

The pgvector extension facilitates the storage and effective searching of embeddings that capture the semantic meaning of text inputs. By utilizing pgvector, customers across all industries can benefit from the power of  Generative AI to add new capabilities to their Postgres applications. 

 

To get started with pgvector in Azure Database for PostgreSQL Flexible Server.

 

How to enable vector extension in Azure Database for PostgreSQL Flexible Server

 

Concepts-pgvector extension in Azure Database for PostgreSQL Flexible Server 

 

How to Optimize Pgvector in Azure Database for PostgreSQL Flexible Server

 

To get started with pgvector in Azure CosmosDB for PostgreSQL.

 

How to enable vector extension in Azure CosmosDB for PostgreSQL

 

Concepts-pgvector extension in Azure CosmosDB for PostgreSQL

 

How to Optimize Pgvector in Azure CosmosDB for PostgreSQL

 

 

What are embeddings and pgvector?

 

Embeddings are vectors created by generative AI models that represent the semantic meaning of textual data. With the pgvector extension, these embeddings can be stored in PostgreSQL databases, allowing efficient similarity searches.

 

This feature is particularly useful for recommendation systems, where businesses can provide personalized suggestions based on a user's previous interactions or preferences. For instance, a streaming service can leverage pgvector to provide a curated list of film recommendations that closely resemble the one recently watched by a user.

 

Benefits of using pgvector in Azure PostgreSQL

 

Easy Integration: With support for pgvector, integrating ML models into Azure Database for PostgreSQL Flexible Server and CosmosDB for PostgreSQL is straightforward, enabling developers to focus on building ML capabilities without worrying about complex database configurations.

 

Scalability: Azure PostgreSQL Flexible Server and CosmosDB for PostgreSQL allows for seamless scalability, ensuring that databases can handle increasing amounts of data and ML workload as applications grow.

 

Efficient Similarity Searches: By utilizing pgvector, businesses can perform efficient similarity searches to find items that closely match specific patterns or user preferences, enhancing recommendation systems and other AI-driven functionalities.

 

 

To help customers get started with pgvector and Open AI, we’ve authored below two new blogs.

 

Learn how OpenAI's retrieval plugins enhance Azure Database for PostgreSQL and CosmosDB for PostgreSQL, enabling seamless integration with OpenAI workloads. Discover the concept of embeddings, the power of pgvector extension, and the capabilities of the retrieval plugin with PostgreSQL databases on Azure for advanced data interactions.

 

OpenAI/ChatGPT retrieval plugin and PostgreSQL on Azure - Microsoft Community Hub

 

Learn about the potential of how OpenAI's chat plugins can enable developers to automate the design of data models. In this blog, Arda shows an example of shard key selection in Cosmos DB for PostgreSQL using plugin’s that make it easy to design distributed data models with expert knowledge of distributed systems.

 

Idea for GPT Plugin: Citus Shard Key Selection in Postgres - Microsoft Community Hub

 

 

Conclusion:

 

The support for the pgvector extension and Open AI plugins in Azure Database for PostgreSQL Flexible Server and Azure CosmosDB for PostgreSQL empowers developers to simplify Open AI integration and enhance their applications' capabilities. By leveraging embeddings and efficient similarity searches, businesses can provide personalized experiences, improve recommendation systems, and unlock the full potential of AI-enabled applications. Whether it is e-commerce, media, healthcare, or other domains, Azure PostgreSQL Flexible Server and Azure CosmosDB PostgreSQL databases with pgvector enables efficient and scalable ML model integration for a wide range of use cases.

 

Further Reading

 

Azure Cosmos DB for PostgreSQL  

Azure Database for PostgreSQL - Flexible Server

OpenAI | Prompt Design

OpenAI | Fine-Tuning

OpenAI | Chat Plugins

 

 

 

1 Comment
Co-Authors
Version history
Last update:
‎May 25 2023 10:16 AM
Updated by: