Australian AI infrastructure company Firmus Technologies has signed a strategic partnership with US chip company Nvidia to procure and operate Nvidia-powered cloud infrastructure, providing emerging and smaller AI firms with more affordable access to high-performance computing capacity.

The deal, reported by Channel News Asia, will see Firmus purchase Nvidia infrastructure and sell Nvidia-powered cloud services to AI-native customers, with Nvidia earning both product revenue and a share of cloud revenue under the arrangement.

The partnership will deliver 170,000 Graphics Processing Units from the first quarter of 2027 through to early 2028, located in Batam, Indonesia. Firmus said it expected to earn up to $30 billion (approximately €25.2 billion) in revenue during the first six years of the deal, based on existing customer commitments.

Tim Rosenfield, co-CEO of Firmus Technologies, said: "We have worked to figure out how to close the gap between the cost benefits that the large guys have access to, which they do because they have great credit ratings, and the guys that are up and comers. This is actually a really material way to level the playing field a little bit to give the next a chance to compete with the big guys."

Nvidia has participated in previous capital raisings by Firmus, making it an existing investor in the Australian company. Firmus raised $1.35 billion (approximately €1.13 billion) over the six months to April 2026, achieving a post-money valuation of $5.5 billion (approximately €4.6 billion). The company has appointed investment banks to explore a potential initial public offering, according to sources cited in the report.

The partnership addresses a structural challenge in the AI infrastructure market, where access to large-scale GPU capacity has historically favoured well-capitalised hyperscalers and large technology companies over smaller and developing AI firms seeking to compete at scale.

Access the full report for further detail on the Firmus Technologies and Nvidia partnership terms and GPU deployment timeline.