Blockmate Ventures Buys NVIDIA Node for AI Compute Test
The firm enters the GPU compute market to pilot performance metrics ahead of a potential expansion in 2027.
Updated on Oct. 27, 2026 in Artificial Intelligence
Live Poll
Would you consider investing in companies focused on building AI infrastructure?
Blockmate Ventures has entered the AI GPU compute market by purchasing an NVIDIA B200 AI compute node from Scannit Inc. for US$635,000. The hardware is expected to be delivered and installed in a U.S. data center in the first quarter of 2027.
Why it matters
This deployment marks Blockmate's transition into the neocloud sector as it seeks to gather operational data on GPU demand and utilization. The pilot serves as a test case for potential future multi-node hardware investments.
The B200 node includes eight GPUs. Blockmate expects to recoup the initial capital investment in roughly two years, though actual performance metrics regarding uptime and pricing remain subject to market volatility.
The players
Blockmate Ventures
An investment firm expanding into the AI neocloud and GPU compute market.
Scannit Inc.
A provider that operates GPU marketplaces and manages hardware for direct contracts.
Hivello
An entity in which Blockmate holds a 50% stake, currently associated with C$7 million in SAFE note liabilities.
Carosa Corporation
A firm currently finalizing a secured loan to assist Blockmate in financing its hardware purchase.
The details
The NVIDIA B200 node consists of a high-performance cluster of eight GPUs (Graphics Processing Units — specialized circuits designed to accelerate the mathematical operations required for AI training and inference). Scannit Inc. will handle the physical data center installation, network connectivity, and ongoing maintenance. Blockmate Ventures retains the majority of gross revenue from the node for a 36-month term, with the hardware itself remaining under the technical oversight of the host provider.
Timeline
June 30: Date of balance sheet listing for Hivello SAFE notes.
Q1 2027: Expected delivery and installation date of the compute node.
The Tech Race
This acquisition aligns with the broader push to expand high-density GPU capacity within the U.S. infrastructure. The project follows the trend of regional data center expansion efforts in Wyoming designed to accommodate large-scale compute clusters.
Investors and industry analysts in Wyoming should watch the Q1 2027 delivery window to confirm the project's operational viability. The firm plans to use this node as a benchmark to determine whether to scale into larger multi-node deployments.
The takeaway
This deployment marks a shift in how smaller firms enter the high-compute market by leasing managed infrastructure. Stakeholders should track the two-year capital recovery projection against actual uptime performance once the node comes online in 2027.
Further reading
Learn more about the infrastructure and hardware trends currently shaping the Artificial Intelligence sector.
Live Poll
Would you consider investing in companies focused on building AI infrastructure?
