Kubernetes operations and cost management startup Cast AI Group Inc. today announced several new features for its platform to reduce cloud costs for organizations that are using generative artificial intelligence models.

The new features from CAST AI are designed to assist organizations that are building, training and running AI models and applications in the cloud. The company argues that as companies rush to embrace AI, especially the generative AI that’s behind chatbots such as OpenAI LP’s ChatGPT, many have not considered the high compute requirements and resources needed. That’s where the new release from CAST AI steps in.

To support teams that are dealing with AI models, CAST AI has expanded its platform with new features, including automated provisioning. The automatic provisioning feature allows users to select and scale cost-effective graphics processing unit-based machines across Amazon Web Services, Google Cloud and Microsoft Azure.

The new release includes automated decommissioning of GPU instances and the ability to deploy a replacement with more cost-efficient alternatives once the process is completed. Users also now have access to automated optimization of Amazon Inferentia machines employed to execute AI models.

Users of high-performance Graviton processors now have additional access to the CAST AI platform to measure performance and cost balance. The upgraded service also offers automated management of spot instances, with CAST AI identifying the optimal pod configuration for the model’s computation requirements and automatically selecting machines that meet these criteria cost-effectively.

The upgrades are claimed to have delivered cost savings of 76% in beta testing for one of CAST AI’s customers by streamlining their AI workflows. The savings came through reducing time spent managing AI models while also driving better results.

“As AI models continue to grow in size and power, the cost of computation to train and run them also increases,” Bobby Yazdani, founder and partner of Cota Capital Management LLC, one of CAST AI’s investors, said ahead of the news. “By significantly reducing these costs, CAST AI can enable wider experimentation, deployment and adoption of these powerful new technologies.”

CAST AI was previously in the news in March when it raised $20 million in new funding to take advantage of the significant opportunity as the market for cloud-native solutions is experiencing hypergrowth. Investors in the company include Cota Capital, along with Samsung Next, Scale Asia Ventures Pte. Ltd., Florida Funders LLC, Tesonet UAB and DNX Ventures LLC.

Image: CAST AI