Cloud AI building blocks

Easily infuse AI into existing applications, or build entirely new intelligent applications for a large spectrum of use cases, even with limited prior machine learning (ML) expertise.

AI building blocks make it easy to add the humanlike capabilities of sight, language, and conversation to your applications.




You can use a single product or a combination of building block products to build the solution that’s right for you.

Introduction to Cloud AI Building Blocks

Get a 90-second overview of how Google Cloud AI Building Blocks can help your business.

Customers using AI building blocks

At Keller WIlliams we have a huge focus on technology and on innovating to make our real estate franchise stand out to customers. We are planning to use AutoML Vision to identify specific items or elements in agent listing photographs, such as ‘modern’ or ‘granite countertops.’ This will make it easier for our customers to find their perfect home!

Neil Dholakia, Chief Product Officer, Keller Williams
Forbes Meredith Hearst PWC Singapore Airlines Ebay Cardinal Health USA today The Telegraph Marketo

Two types of AI building blocks

There are two types of building blocks: APIs for pre-trained models and AutoML for custom models. These building blocks can be used individually or in combination.


Pre-trained models

Utilize pre-trained APIs for common use cases. Immediately start using our proven datasets without worrying about training models, and access the latest fully trained models as Google makes breakthroughs in AI research.


Custom models

Create custom models for domain-specific use cases. From identifying wear on tires to classifying maps and legal documents, Cloud AutoML products use Google’s state-of-the-art transfer learning and neural architecture search technology to build domain-specific models and increase accuracy for a variety of use cases.

Use cases

See how AI building blocks can be used together to develop robust solutions that meet your business needs.

Related content