We turn stranded or fragmented power assets into globally useful AI compute.

Project Ohm is an Australian distributed AI infrastructure and orchestration company. We deploy modular compute nodes where power is already available, at or near substations and at renewable sites, and run them as one network with our own orchestration platform.

Our nodes sit at or near distribution substations and utility sites, and at renewable generation sites. The orchestration platform decides when and where each job runs, weighing energy availability, job urgency, latency, and data security and sovereignty. So the compute rises with the midday surplus and steps back at the peak, and the grid can schedule the whole fleet as one resource. See the platform

AI and high-performance compute that the electricity system can schedule across time and location. Each node connects like a load, and the platform dispatches the fleet like a resource: it soaks up surplus power, steps back on cue and is built to follow a network or market signal. Every node is metered, so network operators see what it drew, when, and how it responded. See it on the grid

Compute customers who want options. The platform weighs each workload's urgency and latency, so flexible jobs follow the cheapest power and time-critical inference runs close to where it's needed. Network operators and energy partners, for demand that rises with the midday surplus and steps back at the peak. And owners of sites with power to spare, such as substations with spare capacity at midday and curtailed wind and solar farms.

Flexible work, such as training, fine-tuning, batch inference, embeddings, synthetic data, simulation and rendering, follows the cheapest and most abundant power: the platform runs it where and when power costs least, and it saves its progress as it runs, so a pause loses nothing. Time-critical work, such as live inference, runs on the node closest to its users.

Each node is a modular, transportable compute unit of 1 to 20 MW, sized to the capacity on site and built with tier-1 hardware partners. It goes in on a nine-month target deployment cycle, using network capacity that already exists rather than waiting years for a large new connection. When a substation no longer needs the extra load, its node goes on a truck to one that does.

Project Ohm rollout plan, September 2026

Nodes and the platform are hardware-agnostic by design: each node carries high-density accelerator racks from any vendor, chosen to suit its work, cooled by air or closed-loop liquid cooling. XENON, a strategic investor, is our long-term hardware supplier.

It starts with Node 01, our first commercial-scale deployment, being developed in Geraldton, WA, on the Western Power network, and a second site is in planning on the east-coast market. Our Australian roadmap targets 600 MW across 300 to 400 substations on the SWIS and the NEM, then Southeast Asia and beyond. See the rollout

Project Ohm rollout plan, September 2026

Our nodes use air cooling or closed-loop liquid cooling, designed to avoid the large ongoing water use of the evaporative cooling many data centres rely on. See it in the community

No large new sites. Our nodes sit at or near existing substations and utility sites, and at renewable generation sites. They connect to the network that is already there, with no new high-voltage lines, and can be relocated if circumstances change. See it in the community

No. Because the work can move, our nodes are designed without diesel or gas generators. A small battery gives running jobs time to save their progress if the power drops. See it in the community

No. A node is built to use power when it is plentiful and step back when it is scarce: it ramps up with the midday surplus and eases off at the evening peak, when households need the power most. See it in the community

Investment, and compute close by. Our Australian roadmap targets 300 to 400 substations, so the investment reaches hundreds of communities rather than one campus. Time-critical work runs on the node nearest its users, so start-ups and researchers nearby get fast Australian compute. See it in the community

Project Ohm rollout plan, September 2026

No. The fleet grows two ways: nodes Project Ohm funds itself, and nodes funded with partners, from utilities and energy partners to infrastructure investors. Beyond the fleet, the orchestration platform is designed to take on third-party compute and schedule it the same way.

Qualified investors can request the investor briefing, and someone from the team will reply in person. Information about investing in Project Ohm is made available on request only to investors who qualify as wholesale, professional or sophisticated investors under Australian law. Read the Investors page

Send a message through the contact form, choosing investing, partnering or something else, and someone from the team will reply. Our email addresses and LinkedIn are at the foot of every page. Contact the team