GMI Cloud Raises $668M for AI Compute

GMI Cloud’s latest financing shows how quickly AI infrastructure has moved from venture capital story to large-scale capital expenditure problem.

What happened

GMI Cloud raised $668 million in new financing, made up of a $223 million Series B equity round and a $445 million credit facility. The company plans to use the capital to expand GPU capacity across the US, Taiwan and Southeast Asia, while growing its inference services.

The round combines venture-style growth capital with debt financing, reflecting the cost of building AI compute capacity. Specialist AI clouds need to secure GPUs, networking equipment, data-centre access, power and customer demand at the same time.

Why it matters

Neocloud expansion increasingly looks like infrastructure finance rather than normal software fundraising. Equity alone is often not enough to buy the physical assets required to compete with hyperscalers, especially when the market is demanding faster deployment and cheaper inference.

That changes how investors should think about the sector. These companies may scale quickly, but they also carry asset-heavy risks around utilisation, debt, hardware depreciation and power availability.

The bigger picture

AI compute is becoming an infrastructure market with venture upside and industrial balance-sheet pressure. The next phase of competition will not only be about who has the most GPUs, but who can finance, operate and fill that capacity efficiently.