Oracle has expanded its partnership with AMD, deploying 50,000 AI GPUs, challenging Nvidia’s dominance

Today, Oracle announced the expansion of its long-standing, multi-generational collaboration with AMD.  Oracle Cloud Infrastructure (OCI) will deploy 50,000 GPUs starting in calendar Q3 2026, with expansion planned for 2027 and beyond.

Oracle and AMD expanded their collaboration to deliver AMD and GPUs on OCI for end customers. The cloud infrastructure firm revealed that the deployment will start with the 2024 AMD Instinct MI300X-powered shapes. Oracle and AMD will extend the general availability of OCI Compute with AMD Instinct MI355X GPUs.

According to Google Finance, shares of AMD rose 3.47% to $223.86, while Oracle’s stock fell 1.85% to $302.33.

Oracle launches new AI superclusters with AMD chips

$ORCL TO DEPLOY 50K $AMD MI450 AI GPUS STARTING IN 2026

It’s one of the largest AMD cloud wins to date & signals Oracle’s push to scale its AI infrastructure beyond $NVDA supply constraints. pic.twitter.com/YVGiayTpae

— Shay Boloor (@StockSavvyShay) October 14, 2025

Oracle stated that demand for large-scale AI capacity is accelerating as next-generation AI models outgrow the limits of current AI clusters. Oracle believes that customers require flexible, open-compute solutions designed for extreme scale and efficiency to train and run these workloads.

“Our customers are building some of the world’s most ambitious AI applications, and that requires robust, scalable, and high-performance infrastructure.” 

–Mahesh Thiagarajan, Executive Vice President, Oracle Cloud Infrastructure

Thiagarajan added that customers will push the boundaries of AI applications with confidence by integrating the latest AMD processors with OCI’s secure platform, powered by Oracle Acceleron.

Forrest Norrod, Executive Vice President and General Manager of the Data Center Solutions Business Group at AMD, said that AMD and Oracle continue to set the pace for AI innovation in cloud computing. Narrod emphasized that, by using AMD Instinct GPUs and EPYC CPUs, Oracle customers will gain new capabilities for training, fine-tuning, and deploying the next generation of AI. He said the two firms are accelerating AI development with their open, secure, and optimized solutions. Narrod also emphasized that the optimized solutions are designed specifically for large AI data centers.

Karan Batta, Senior Vice President of Oracle Cloud Infrastructure, told CNBC’s Seema Mody that he believed customers would adopt AMD, particularly in the inference space.

OpenAI deepens partnership with AMD in expanding AI infrastructure

Earlier this month, AMD and OpenAI signed a contract for processors that require 6 gigawatts of electricity over several years, with a 1-gigawatt deployment scheduled to begin in 2026. AMD has issued OpenAI a warrant for up to 160 million shares of common stock upon the successful deployment of 1 gigawatt.

The deal positions AMD as a core strategic partner to OpenAI, marking one of the largest GPU deployment agreements in the artificial intelligence industry to date.

OpenAI has historically maintained close ties with Nvidia, using its chips to develop ChatGPT. Nvidia’s chips dominate the market for data center GPUs, holding over 90% of the market share. Nvidia also invested in OpenAI in September.

As Cryptopolitan reported on September 23, Sam Altman and Jensen Huang finalized a $100 billion agreement between OpenAI and Nvidia. The agreement solidified the chip giant’s position as the engine behind the upcoming OpenAI models. The arrangement included Long-term hardware supply and capital investment. However, in Nvidia’s case, the chipmaker acquired a share in OpenAI.

OpenAI executives noted that the company requires substantial processing power, which necessitates the use of AI processors from multiple vendors.  

On Monday, OpenAI also announced an agreement with chipmaker Broadcom Inc. The agreement meant to add 10 gigawatts of capacity. OpenAI will design the accelerators and systems, which will be developed and deployed in partnership with Broadcom. Hock Tan, President and CEO of Broadcom, said, Broadcom’s collaboration with OpenAI signifies a pivotal moment in the pursuit of artificial general intelligence. In September, OpenAI and Oracle signed a five-year cloud agreement valued at approximately $300 million.

If you're reading this, you’re already ahead. Stay there with our newsletter.