GMI Cloud, a leading independent AI infrastructure provider in the US, is pursuing $635 million in GPU-backed financing. Nvidia is reportedly supporting the deal. The company operates over 30,000 Nvidia GPUs, including H100, H200, and Blackwell chips, housed in US data centers. It specializes in AI inference workloads, the production phase of artificial intelligence.
GMI Cloud holds a distinguished status as one of only six global Reference Platform Cloud Partners certified by Nvidia. It was also an early contributor to Nvidia's managed cloud AI service, DGX Cloud Lepton. Neither company has publicly disclosed the financing terms, potential lenders, or a closing timeline.
GPU-backed lending is an emerging trend. High-end Nvidia chips carry list prices in the tens of thousands of dollars. A large fleet with active workloads provides lenders a tangible asset base for underwriting, with key variables being utilization rates, customer contracts, and hardware lifespan.
The financing would likely fund additional GPU procurement and data center expansion. This would help GMI Cloud compete against rivals like CoreWeave, Lambda, and major hyperscalers. The deal's structure also aligns with the growing interest in tokenizing real-world assets, such as using physical compute hardware as collateral for blockchain-based financial instruments.
The interest rate and loan-to-value ratio from this financing will signal how lenders price GPU collateral risk, setting a precedent for AI infrastructure companies funding buildouts with debt.