AI Frontier Post
AI News

Boost Run lands $525.6M AI compute deal with an unnamed sovereign AI company — backlog tops $2.6B

Boost Run (Nasdaq: BRUN) has signed a five-year, $525.6 million AI cloud services agreement with a "leading sovereign AI company," pushing the Nvidia-preferred cloud provider's contracted backlog past $2.6 billion as the GPU-cloud land grab intensifies.

Boost Run, Inc. (Nasdaq: BRUN), an NVIDIA Preferred Cloud Partner, said Tuesday it has signed a five-year AI cloud services agreement with a leading sovereign AI company carrying a total contract value of $525.6 million. Acceptance of the initial infrastructure is expected to begin in early Q2 2027. The deal lifts Boost Run's total contracted backlog to more than $2.6 billion in committed customer revenue.

GB300 NVL72, storage, and CPU nodes

Under the agreement, Boost Run will deliver compute capacity powered by NVIDIA GB300 NVL72 GPUs, along with network storage and CPU node services. The press release was issued via Business Wire at 8:00 a.m. ET on Tuesday. TheFly's same-day note confirmed the headline terms: a five-year agreement, $525.6M total contract value, and the $2.6B backlog figure.

Close-up of AI accelerator GPU server blades with copper heat sinks and glowing status LEDs inside a dark server rack.
The deal centers on NVIDIA GB300 NVL72 systems plus network storage and CPU nodes. Image: AI Frontier Post (AI-generated).

The sovereign AI angle

The customer is not named — just a "leading sovereign AI company" — which fits a pattern. Governments and state-backed entities are racing to secure domestic AI compute, and they are increasingly doing it through cloud contracts rather than building everything themselves. Boost Run says its financing structure is designed to deploy infrastructure against contracted demand without relying on equity dilution, and that the company's signed contracts include upfront payments.

"This agreement marks an important step in efficiently deploying our financing into contracted revenue," said founder and CEO Andrew Karos. "Our ability to deliver NVIDIA GB300 systems to leading AI companies demonstrates the strength and diversification of both our customer base and our GPU fleet."

Who is Boost Run

Boost Run describes itself as an NVIDIA Preferred Cloud Provider that has also achieved NVIDIA Exemplar Cloud status on the Blackwell architecture. Its platform offers GPU compute, CPU nodes, managed Kubernetes orchestration, and shared storage through a console and API layer, and it carries SOC 2 Type II, HIPAA, ISO 27001, and ISO 27701 certifications at the operator level. The company says it plans to keep converting its pipeline into signed contracts, deploying capacity against contracted demand.

Aerial dusk view of a large data center campus with glowing buildings under a deep blue sky.
Boost Run says it will keep converting its pipeline into signed contracts as the AI compute market expands. Image: AI Frontier Post (AI-generated).

What to watch

Two things matter now. First, who the sovereign customer is — if it is disclosed, it will say a lot about which state is buying GB300 capacity at this scale. Second, execution: acceptance starts in early Q2 2027, and Boost Run still has to build and deliver the infrastructure behind a headline TCV that now exceeds $2.6 billion. The $525.6M deal is a vote of confidence in the company's ability to do exactly that.