Torus Foundry brings AI inference capacity to existing grid connections
Torus launched Foundry, a factory-integrated system combining energy storage, cooling, networking and GPU compute in deployments ranging from 1 to 20 MW.
AT A GLANCE
Key points
- Foundry combines power and GPU compute in a prefabricated system.
- Deployments range from 1 to 20 MW on existing grid connections.
- Torus targets 12–16 weeks from contract to running infrastructure.
What is changing?
Torus introduced Foundry, a prefabricated system that combines power, energy storage, cooling, networking and GPU compute for enterprise AI inference workloads. Instead of building a new hyperscale campus, the system is installed where commercial and industrial grid connections already have usable capacity. Torus is targeting 12–16 weeks from contract to running workloads.
Foundry can be deployed at 1–20 MW per site and is launching with initial customers including Hydra Host. The modules are pre-tested at the GigaOne facility in Salt Lake City; the company says on-site installation and commissioning take one to two days. A separately announced $500 million growth-capital package will support manufacturing and U.S. deployments. Timing, capacity and efficiency figures are provider targets.
Explore the original sources
Source date: 11 November 2026.
Published by AI Radar: 11 Nov 2026, 17:10 (Istanbul, UTC+3).
FOLLOW THE TOPIC
Topics and tags
Share your discovery.
Send this story to a friend.
Reader comments
What do you think about this development? Share your experience or question. Comments are published after editorial approval.
Your name/nickname and comment become public if approved. Read: Privacy notice.
Loading comments…