OpenAI and Cerebras Forge $10 Billion Partnership to Revolutionize AI Compute
In a landmark move poised to reshape the artificial intelligence (AI) landscape, OpenAI has entered into a multi-year agreement with AI hardware innovator Cerebras Systems. This strategic partnership, valued at over $10 billion, will see Cerebras supplying OpenAI with 750 megawatts of computational power annually from 2026 through 2028. This substantial infusion of compute resources is set to significantly enhance OpenAI’s capabilities, enabling faster and more efficient AI model processing.
Accelerating AI Performance
The collaboration between OpenAI and Cerebras is driven by a shared commitment to delivering rapid and reliable AI outputs. OpenAI has emphasized that integrating Cerebras’ advanced systems will expedite response times for complex queries, thereby improving user experience and interaction. Andrew Feldman, co-founder and CEO of Cerebras, drew a parallel to the transformative impact of broadband on the internet, stating, Just as broadband transformed the internet, real-time inference will transform AI.
Cerebras’ Ascendancy in AI Hardware
Founded over a decade ago, Cerebras has emerged as a formidable player in the AI hardware sector, particularly following the AI surge initiated by the launch of ChatGPT in 2022. The company’s unique architecture, centered around its Wafer-Scale Engine (WSE) chips, is specifically designed to outperform traditional GPU-based systems, such as those offered by Nvidia. This design enables faster processing speeds and more efficient handling of AI workloads.
Despite filing for an initial public offering (IPO) in 2024, Cerebras has postponed its public debut multiple times. In the interim, the company has successfully secured substantial funding, including recent discussions to raise an additional $1 billion at a valuation of $22 billion. Notably, OpenAI’s CEO, Sam Altman, is an investor in Cerebras, and there have been considerations in the past for OpenAI to acquire the hardware firm.
Strategic Implications for OpenAI
This partnership aligns with OpenAI’s strategic objective to diversify and fortify its computational infrastructure. By incorporating Cerebras’ low-latency inference solutions, OpenAI aims to deliver faster responses and more natural interactions, thereby scaling real-time AI applications to a broader audience. Sachin Katti of OpenAI articulated this vision, stating, Cerebras adds a dedicated low-latency inference solution to our platform. That means faster responses, more natural interactions, and a stronger foundation to scale real-time AI to many more people.
Broader Context in AI Infrastructure Investments
OpenAI’s agreement with Cerebras is part of a broader trend of substantial investments in AI infrastructure. In October 2025, OpenAI partnered with Broadcom to develop custom AI accelerator hardware, committing to 10 gigawatts of compute capacity to be deployed between 2026 and 2029. This deal, estimated to cost between $350 billion and $500 billion, underscores OpenAI’s dedication to building a robust and scalable AI infrastructure.
Additionally, in October 2025, OpenAI entered into a multi-year agreement with AMD to supply 6 gigawatts of compute capacity across multiple generations of AMD’s Instinct GPUs, starting with the Instinct MI450 GPU. This partnership, valued in the tens of billions, is expected to enhance OpenAI’s computational capabilities significantly.
In July 2025, OpenAI agreed to pay Oracle $30 billion annually for data center services as part of the Stargate project, a $500 billion data-center-building initiative. This deal involves 4.5 gigawatts of capacity, equivalent to the power needed for approximately four million homes, highlighting the scale of OpenAI’s infrastructure expansion efforts.
Conclusion
The $10 billion partnership between OpenAI and Cerebras Systems marks a significant milestone in the evolution of AI infrastructure. By leveraging Cerebras’ cutting-edge hardware solutions, OpenAI is poised to enhance the speed and efficiency of its AI models, delivering more responsive and natural interactions to users. This collaboration not only underscores the growing importance of specialized hardware in AI development but also reflects the broader trend of substantial investments aimed at scaling AI capabilities to meet increasing global demand.
