Firmus and OpenAI have announced a multi-year strategic partnership under which OpenAI will contract dedicated artificial intelligence compute capacity from two Firmus AI Factory sites in Malaysia. The agreement, dated 8 September 2026, makes OpenAI an anchor customer of Firmus and supports the deployment of additional AI factory capacity across the company's AI Factory platform. The collaboration extends Firmus' regional reach and strengthens its position in the fast-growing Asia-Pacific AI infrastructure market.
Contracted Capacity and Regional Footprint
With this partnership, Firmus' total contracted capacity across all customers now exceeds 900 megawatts, underpinning the next phase of the company's growth across the region. The broader portfolio spans seven AI factories across Australia, Singapore, Indonesia and Malaysia. Two sites are operational today in Australia and Singapore, while five others are under development and targeting ready-for-service within the next 24 months.
Malaysian Expansion and Regional Demand
Malaysia becomes a new market for Firmus, extending the company's Asia-Pacific footprint beyond its existing presence in Australia, Singapore and Indonesia. The two dedicated AI factory sites in Malaysia are expected to support OpenAI's regional and global compute requirements. This expansion aligns with growing enterprise and developer demand for advanced AI services across Southeast Asia and the wider region.
NVIDIA Vera Rubin and HyperCube Integration
Firmus will deploy NVIDIA Vera Rubin at scale across Asia-Pacific, bringing the next-generation accelerated computing platform to the region. The AI factories developed under the partnership will use the NVIDIA DSX AI Factory Platform to integrate Vera Rubin NVL72 rack-scale systems with Firmus' HyperCube platform. The HyperCube system combines advanced liquid cooling, mechanical systems and electrification into a repeatable infrastructure solution manufactured and prefabricated in regional New South Wales.
Australian Engineering and Prefabrication
Firmus' HyperCube infrastructure is manufactured and prefabricated in regional New South Wales, supporting local engineering capacity and supply chain activity. The integration of power, cooling and mechanical systems into repeatable modules is designed to reduce deployment timelines and improve consistency. This Australian engineering base underpins the company's ability to scale AI factory capacity rapidly across multiple countries.
Operational Efficiency and Token Economics
Bringing more of the AI factory value chain together is expected to improve deployment speed, capital efficiency and operating performance. Those gains are intended to support a lower cost of token production for customers and developers. The partnership also reflects a broader industry shift toward integrated AI factories that span power, cooling, accelerated computing systems, networking, software and workload performance.
Executive Perspectives on Regional AI Growth
Firmus Co-Founder and Co-CEO Tim Rosenfield said the partnership marks the moment Asia-Pacific becomes a producer of intelligence rather than only a consumer. OpenAI Vice President of Compute Strategy Sachin Katti said the Malaysian data centers will help serve rising demand for OpenAI products across the region and globally. NVIDIA Vice President Nico Caprez added that the HyperCube-based AI factories should help bring advanced capacity online faster with stronger token economics.
Australian AI Access Program
Alongside the OpenAI partnership, Firmus also intends to establish an Australian AI Access Program to support eligible researchers and organisations. The proposed program will focus on applications of artificial intelligence to areas of national and regional importance. Key focus areas are expected to include science, education, agriculture, energy and resource efficiency, and climate resilience.
The strategic partnership positions Firmus at the centre of Asia-Pacific AI infrastructure growth by combining OpenAI-scale demand with NVIDIA technology and Australian manufacturing. It also demonstrates how regional AI factory platforms can support global compute needs while pursuing lower token production costs. The collaboration is expected to accelerate advanced AI capacity across Southeast Asia and Australia over the coming years.