Firmus Signs OpenAI as Anchor Customer, Crosses 900 MW Contracted Capacity and Enters Malaysia

project_news

Firmus Signs OpenAI as Anchor Customer, Crosses 900 MW Contracted Capacity and Enters Malaysia

Updated on Sep 08, 2026, 05:36 PM IST
Written & Edited by Ashish

Firmus Technologies has secured OpenAI as an anchor customer under a multi-year strategic partnership, pushing the Australian AI infrastructure company's total contracted capacity past 900 megawatts and extending its Asia-Pacific footprint into Malaysia.

Partnership Centres on Dedicated Malaysian AI Compute

The agreement, announced in Sydney on 8 September 2026, will see OpenAI contract dedicated AI compute capacity from two Firmus AI Factory sites in Malaysia.

 

OpenAI's role as an anchor customer is intended to support the deployment of additional AI Factory capacity across Firmus' broader platform. The company said the contracted customer portfolio as a whole underpins the next phase of its regional growth.

Sachin Katti, Vice President of Compute Strategy at OpenAI, said the company was looking to expand its global compute capacity and that the Malaysian data centres would help serve growing demand for OpenAI's products across the region and around the world. Katti said OpenAI looked forward to bringing the capacity online alongside Firmus.

 

A Seven-Country Footprint Built Over Seven Years

The Malaysia announcement extends Firmus' presence beyond its existing operations in Australia, Singapore, and Indonesia. The company's broader portfolio now spans seven AI factories across four countries.

 

Two sites are currently operational, one in Australia and one in Singapore, while five additional sites are under development, with all five targeting a ready-for-service date within the next 24 months.

Tim Rosenfield, Co-Founder and Co-CEO of Firmus, described the OpenAI partnership as the moment Asia-Pacific becomes a producer of intelligence rather than just a consumer of it.

 

He said the company had spent seven years building toward this point, developing a footprint running from Southeast Asia to Australia along with the manufacturing, technology, and teams required to bring it online at speed.

Trusted by Leading EPCs & Manufacturers

Find the Latest Data Center Projects in Malaysia

Gain exclusive access to our industry-leading database of data center opportunities with detailed project timelines and stakeholder information

Request Free Trial → Learn More →

No credit card Up-to-date coverage

 

NVIDIA Vera Rubin at the Centre of the Technology Stack

The AI factories being developed under the partnership will deploy NVIDIA's Vera Rubin platform at scale across Asia-Pacific. Specifically, the sites will use the NVIDIA DSX AI Factory Platform to integrate NVIDIA Vera Rubin NVL72 rack-scale systems with Firmus' proprietary HyperCube platform.

Firmus describes HyperCube as a system that integrates advanced liquid cooling, mechanical systems, and electrification into a repeatable infrastructure unit. The company manufactures and prefabricates HyperCube in regional New South Wales.

 

Firmus said bringing more of the AI factory value chain together in this way is expected to improve deployment speed, capital efficiency and operating performance, ultimately supporting a lower cost of token production.

Nico Caprez, Vice President of Global AI Infrastructure Growth at NVIDIA, said AI infrastructure demand was accelerating across Asia-Pacific and that meeting it required full-stack AI factories engineered for performance, durability, efficiency, scale and fungible workloads.

 

Caprez said the HyperCube-based AI factories built with NVIDIA DSX and powered by NVIDIA Vera Rubin NVL72 systems would help bring advanced capacity online faster and produce intelligence at industrial scale with stronger token economics across the region.

Vertical Integration From Model to Grid

Firmus positions itself as vertically integrated from model to grid, using proprietary energy and cooling technology to build and deliver what it calls next-generation infrastructure.

 

The company is headquartered in Sydney and provides AI cloud services across Asia-Pacific and beyond. It trades as Firmus Technologies under the registered entity Firmus Grid Limited.

The company's stated model is centred on the concept of energy in and tokens out, with efficiency as the organising principle across the entire stack. The HyperCube manufacturing operation in regional New South Wales is a central element of that model, allowing Firmus to control a significant portion of the physical infrastructure supply chain within Australia rather than relying entirely on imported systems.

Australian AI Access Program Proposed Alongside Commercial Expansion

In addition to the commercial partnership with OpenAI, Firmus said it intends to establish an Australian AI Access Program designed to support eligible researchers and organisations applying AI to areas of national and regional importance.

 

The proposed program is expected to focus on applications in science, education, agriculture, energy and resource efficiency, and climate resilience. Firmus described the program as a separate initiative running alongside its commercial expansion rather than a component of the OpenAI partnership itself.

Milestone Comes After a Series of Rapid Moves

The 900-megawatt contracted capacity figure and the OpenAI partnership follow a period of significant activity for Firmus. Earlier milestones referenced in the company's newsroom include a June 2026 announcement about plans to build a 170,000-GPU AI factory campus with NVIDIA for global AI-native customers, a July 2025 announcement about a world-first AI Factory Zone designation in Tasmania clearing the path for a project called Southgate, and a November 2025 capital raise to expand Project Southgate nationally.

 

The Malaysia expansion and the OpenAI anchor agreement represent the latest additions to a pipeline the company has been assembling across multiple years and multiple jurisdictions.

Track Every Megawatt of Malaysia's Data Center Boom Before Your Competitors Do

Malaysia is adding data center capacity at a pace that few markets have ever seen, and the window to position your business ahead of major procurement and partnership opportunities is narrowing fast. The organizations winning contracts in this space are not simply reacting to announcements; they are working from intelligence that gives them a structural head start.

 

The Global Project Tracking (GPT) platform by Blackridge Research was built precisely for this environment. It consolidates the fragmented landscape of data center development activity across Malaysia into a single, continuously updated intelligence resource, giving your team the visibility needed to act with confidence rather than guesswork.

 

Whether you supply cooling infrastructure, fiber connectivity, power systems, or construction services, the depth of project data available through the platform means you can align your pipeline strategy with where real spending is actually headed across the region.

 

  • Upcoming Projects

  • Tender Notices

  • Contract Awards

  • Projects Under Construction

  • Completed Projects

 

See the full scope of Malaysia data center activity mapped to your business needs. Book a Free Demo with the Blackridge Research team today.

Tags

Leave a Comment

We love hearing from our readers and value your feedback. If you have any questions or comments about our content, feel free to leave a comment below.

We read every comment and do our best to respond to them all.

Protected by Cloudflare Turnstile