Google Cloud's Vertex AI, made generally available on May 18, provides a managed machine learning platform for the deployment and maintenance of artificial intelligence models. Google says the Vertex AI platform requires fewer lines of code to train a model than other systems.
Vertex AI unites all Google Cloud services for building machine learning models under a unified UI and API, simplifying the process of building and deploying machine learning models at scale, Google said. Specifically, AutoML and AI Platform are tied together into a unified API, client library, and UI.
Users can manage data and prototype, deploy, and interpret models without needing formal machine learning training, the company said.
Specific capabilities of Vertex AI include accessing the Google AI toolkit powering Google internally, including pre-trained APIs for computer vision, video, natural language, and structured data.
This is in addition to faster deployment of AI applications via MLOps features such as Vertex Vizier, to increase the rate of experimentation; Vertex Feature Store, to serve, share, and reuse machine learning features; and Vertex Experiments, to accelerate deployment of models.
In addition, tools such as Vertex Continuous Monitoring and Vertex Pipelines streamline machine learning workflow. These tools are intended to remove the complexity of self-service model maintenance and repeatability.