neocloud · vm · Layer 2.5
Massed Compute
Single-node work across a wide spread of NVIDIA parts, from a workstation card to an H200, where a quick start and no commitment matter more than a cluster.
medium portability Movable with effort. Expect to redo images, storage wiring and networking.
What you are buying
- Business model
- Purpose-built GPU cloud. Owns or leases its own datacentre capacity and sells it directly.
- How you reach it
- You get virtual machines. The familiar cloud model.
- Accelerators
- NVIDIA H100 (PCIe, NVL and SXM5), H200 NVL, A100, L40S, L40, RTX 6000 Ada, RTX PRO 6000, RTX PRO 4500, RTX A6000, A5000 and A30 on its price page; B200 and B300 on request.
- Regions
- Not published on its price page. Its August 2025 funding release describes Tier III datacentres and customers across three continents.
- Pricing model
- Hourly on demand, published per configuration from one GPU up, with no long-term commitment. B200 and B300 by quote.
- Getting started
- Self-serve on its own console; it says a machine deploys in under 4 minutes.
- Capacity
- Not published as a GPU count.
- Ownership
- Private. Digital Alpha committed up to $300M in equity and revenue-share financing (HPCwire, 12 Aug 2025).
Verified 2026-11-17. Fields reading “not published” are exactly that — we do not estimate a figure a vendor withholds.
The counterparty
Its price page says it owns and operates all its infrastructure, so you sign with the company that holds the hardware. Part of its financing is a revenue share, so its backer takes a share of rental income; the release names no customers.
A multi-year GPU commitment is a credit decision wearing a cloud contract. This is the section no benchmark covers and the one that decides what happens to your workload in 2028.
The catch
It does not publish where its machines are, so a data-residency requirement means asking before you rent. Blackwell (B200 and B300) is quote-only.
The economics
More than a dozen NVIDIA configurations on one hourly page, with no commitment, from a company that owns the machines. That makes it a useful check on what a single GPU should cost.
No rate is quoted on this page on purpose. Published list prices at this layer move weekly and essentially nobody signing a real contract pays them. Dated, sourced figures live in the GPU cost calculator, where the cheapest and dearest published on-demand price for the identical H100 sit 5.6× apart.
The layers underneath this one
Whatever you rent here is a chip in a building that needs power. In 2026 megawatts, not silicon, are the binding constraint on the whole industry — a frontier rack draws 120–200 kW against a 2026 average near 27 kW, and the US interconnection queue exceeds 2,600 GW.
Layer 2 — Silicon · Layer 1 — Energy · The interconnection queue · Tokens per watt
Related providers
- Crusoe — neocloud, medium portability
- DataCrunch — neocloud, medium portability
- Genesis Cloud — neocloud, medium portability
- Hyperstack (NexGen Cloud) — neocloud, medium portability
- Lambda — neocloud, medium portability
- Lightning AI — neocloud, medium portability