Home > Article > Technology peripherals > Google releases AlloyDB AI to help developers build generative AI applications
Google issued a press release on August 31, Beijing time, announcing the launch of the AlloyDB AI service to further expand the existing AlloyDB service capabilities. This service is designed to help developers make full use of operational data and create generative AI applications
AlloyDB is a fully managed database service compatible with PostgreSQL (an open source database). It automates management tasks such as backup, replication, patching, and capacity management, and leverages adaptive algorithms and machine learning techniques for PostgreSQL vacuum management, storage and memory management, data tiering, and analytics acceleration. In this way, developers can focus on building applications, which is more convenient
The AlloyDB AI introduced this time allows developers to combine the functions of large-scale language models and real-time operation data to provide comprehensive support for vector embedding. AlloyDB AI uses simple SQL functions to convert data into vector embeddings and can perform vector queries up to ten times faster than standard PostgreSQL
AlloyDB AI provides an easy way for users to generate embedded data using PostgreSQL functions. Officials emphasize that with just one line of SQL syntax, you can access Google's embedded models, such as embedded generation models in the database or various final models of Vertex AI
The above is the detailed content of Google releases AlloyDB AI to help developers build generative AI applications. For more information, please follow other related articles on the PHP Chinese website!