BUZZ HPC's 320 MW sovereign AI facility in the Greater Toronto Area signals a major bet on Canadian-built AI infrastructure
Canada has been talking about building domestic AI infrastructure for years. HIVE Digital Technologies is apparently done waiting for someone else to do it.
On May 18, 2026, HIVE announced plans for a 320 MW sovereign AI gigafactory in the Greater Toronto Area, operating through its wholly owned subsidiary BUZZ High Performance Computing Inc. The facility is projected to deploy over 100,000 GPUs upon completion, with an estimated price tag of approximately CAD $3.5 billion and a target operational date in the second half of 2027.
What BUZZ HPC is actually building #
The gigafactory is designed for the Toronto-Waterloo corridor. HIVE’s Executive Chairman Frank Holmes framed the project as foundational to Canada’s AI future.
Key partners include NVIDIA and Hypertec. The facility is being designed with a power usage effectiveness rating below 1.3, with advanced cooling technologies and renewable energy sourcing central to the design.
HIVE is publicly traded on both the Toronto Stock Exchange and NASDAQ under the ticker HIVE.
The Bell contract and the current runway #
On June 18, 2026, BUZZ HPC secured a three-year contract with Bell AI Fabric valued at USD $220 million. The contract covers the deployment of 2,304 NVIDIA Grace Blackwell GPUs at Bell’s facility in Merritt, British Columbia.
USD $220 million over three years works out to roughly USD $73 million annually in committed revenue. BUZZ HPC currently operates 5,500 GPUs for AI compute.
HIVE’s broader expansion roadmap targets approximately 130,000 GPUs across multiple locations. The company’s first GPU cluster, a 504-GPU installation in Manitoba, launched in February 2026.
Why this matters beyond Canada #
The Toronto-Waterloo corridor houses the Vector Institute, the University of Waterloo’s AI programs, and a dense cluster of AI startups.
For investors, the key risk is execution timing. CAD $3.5 billion in capital expenditure is substantial for a company of HIVE’s current size, and the 2027 operational target depends on permitting, construction, and GPU supply chains all moving on schedule. The 130,000 GPU target across all locations would represent a significant operational leap from today’s 5,500-GPU base. Disclosure: This article was edited by Editorial Team. For more information on how we create and review content, see our