Oracle announced plans to integrate Advanced Micro Devices’ (AMD) upcoming MI450 artificial intelligence chips into its cloud infrastructure, with deployment scheduled to begin in the third quarter of 2026. The companies said the initial rollout will include 50,000 processors, with further expansion expected through 2027 and beyond.
The partnership marks a major win for AMD, securing another top-tier client for its next-generation AI chips, while giving Oracle Cloud Infrastructure (OCI) a competitive boost in the global race to provide compute power for AI model training and deployment. “Demand for large-scale AI capacity is accelerating as next-generation AI models outgrow the limits of current clusters,” the companies said in a joint statement.
The announcement comes as demand for AI hardware surges amid the explosion of applications like ChatGPT. AMD’s shares rose over 3% in premarket trading, defying broader market weakness driven by renewed U.S.-China trade tensions, while Oracle’s stock slipped about 1%.
AMD recently unveiled a multi-year deal with OpenAI to supply the same MI450 chips, in an agreement that gives the ChatGPT developer an option to acquire up to 10% of AMD. The companies are also collaborating on a 1-gigawatt AI data facility based on the chip architecture.
The new AI superclusters at Oracle will use AMD’s “Helios” rack design, a fully integrated system combining GPUs and CPUs, mirroring Nvidia’s own rack-scale solutions. The deal underscores AMD’s ambition to challenge Nvidia’s dominance in the high-performance AI hardware market.