Nvidia's enterprise-grade artificial intelligence (AI) software and supercomputing platform are now available on Oracle Cloud Infrastructure (OCI) in a move to provide joint capabilities that will give Oracle a foothold in the growing AI market.

With Nvidia AI Enterprise and Nvidia DGX Cloud on OCI, enterprises can access Nvidia's end-to-end AI development platform and deploy generative AI (genAI) applications on Oracle's cloud infrastructure. IT teams can customize and train their model, then use the same software containers and tools to deploy inferencing on any OCI instance using Nvidia's Triton Inference Server.

Oracle VP Leo Leung told SDxCentral he expects these joint offerings will both draw in new customers and be popular with existing OCI enterprise users, who can integrate their current OCI workloads with Nvidia's technology. "The market for AI is booming, and Nvidia and Oracle are partnering to provide unique capabilities together to attract new customers," Leung said.

Qualified customers – or those who require "clusters of OCI’s powerful bare metal compute instances" as opposed to a single OCI bare metal or VM instance, which is enough for most AI customers to get the job done – can use existing Oracle Universal Credits to purchase access to these Nvidia technologies.

OCI is particularly suited for high-performance computing (HPC) thanks to the nature of the cloud's elasticity and consumption-based costs, which offer "the potential to scale tens of thousands of cores simultaneously," Leung said. "Customers gain access to high-frequency processors; fast and dense local storage; high throughput, ultra-low latency RDMA cluster networking; and tools to automate and run jobs seamlessly."

Gemelo.ai, for example, is running Nvidia AI software development kits and cloud-native microservices on Oracle's cloud infrastructure to power text-to-speech, voice-to-voice and voice cloning systems that can recreate anyone's likeness to act as trained, interactive customer service agents – think Meta's AI assistants based on personas of celebrities like Snoop Dogg and Kendall Jenner.

"We are excited to put the dual resources of OCI and the Nvidia AI Enterprise suite to use in building our next-generation AI-driven applications and ever more useful digital twins," Gemelo.ai CEO Paul Jaski said.

And while this type of HPC is inevitably investment-heavy, "customers are able to combine Nvidia’s powerful software with OCI’s unique capabilities, including bare metal compute with GPUs and cluster networking at supercomputer scale, all at exceptional price-performance," Leung said.