AMD and OpenAI Join Forces to Build the Future of Global AI Infrastructure with 6GW of AMD GPUs
AMD and OpenAI have officially announced a strategic partnership that will position AMD technology at the core of OpenAI’s next-generation AI infrastructure. Under this multi-generational agreement, OpenAI will deploy AMD GPUs with a total power capacity of 6 gigawatts to support large-scale AI operations. The first 1GW phase is set to begin installation in the second half of 2026, utilizing the cutting-edge AMD Instinct MI450 series.
Read the full press release at:Â https://ir.amd.com/news-events/press-releases/detail/1260/amd-and-openai-announce-strategic-partnership-to-deploy-6-gigawatts-of-amd-gpus.
This long-term collaboration marks a pivotal shift in the AI landscape, establishing AMD as a primary compute partner for OpenAI and enabling rapid expansion of its global infrastructure. It is expected to generate tens of billions of dollars in annual AI-related revenue for AMD while accelerating OpenAI’s capacity to deliver next-level AI models and services.
Key highlights of the partnership include a multi-year, multi-generation agreement to deploy up to 6GW of AMD Instinct GPUs that will power OpenAI’s future infrastructure. The deal reinforces AMD’s leadership in open and scalable high-performance AI computing and includes performance-based warrants allowing OpenAI to purchase up to 160 million AMD shares if specific goals are met. The first deployment phase will roll out in 2026 with MI450 GPUs and advanced rack-scale systems designed for large-scale AI workloads.
This announcement solidifies AMD’s position as a leader in the AI infrastructure market, building on its proven track record in data center GPU technology, open-source ROCm software, an extensive open ecosystem, and rack-scale systems engineered for global AI computing demands.
