Hyperstack

Self-serve GPU cloud from UK-based NexGen Cloud

Mid-size cloudEuropeInference

Hyperstack is the self-serve GPU cloud of London-based NexGen Cloud, with regions in Norway, Sweden, Canada and the US. It publishes on-demand, spot and reservation prices, including B200 at $6.00 and H100 SXM at $3.20 per GPU-hour on demand, and reservations from $2.72 for H100 SXM. It holds SOC 2 Type II and is rated Bronze in SemiAnalysis ClusterMAX 3.0.

Best for: Low published on-demand B200 and H100 prices from a ClusterMAX-medaled cloud.

ClusterMAX 3.0: Bronze

✓ Publishes prices
✓ Serverless inference
– AMD GPUs
✓ European data centers

Visit Hyperstack

Facts checked November 9, 2026 against the sources listed below. Hyperstack does not pay to appear here.

1 Legal nameNexGen Cloud Ltd

2 HeadquartersLondon, United Kingdom

3 Founded2020

4 Funding or listingPrivate. NexGen Cloud $45M Series A (Apr 2025); $34M USD.AI facility to finance B200s in Sweden (Jun 2026).

5 How you rentOn-demand, Reserved clusters, Serverless inference

6 Minimum commitmentNone for on-demand (per-minute, prepaid); reservations and Secure Private Cloud are custom-scoped

7 InterconnectOn-demand: SR-IOV networking up to 350 Gbps (CANADA-1, US-1); Secure Private Cloud on NVIDIA Quantum InfiniBand

8 OrchestrationKubernetes (managed), Terraform, SDKs (Go, Python)

9 GPUs offeredGB200 NVL72, GB300 NVL72, B300, B200, RTX PRO 6000, H200, H100 SXM, H100 PCIe, A100, L40, RTX A6000, RTX A4000

10 RegionsNorway (NORWAY-1), Sweden, Canada (CANADA-1, Quebec), US (US-1, Texas)

11 CertificationsSOC 2 Type II

12 Customers and partnersNVIDIA Preferred Partner; announced partnerships with InFlux Technologies, Hugging Face (AI Studio integration) and Utilidata.

Published prices

On-demand list prices per GPU-hour, as published on November 9, 2026. Reserved and negotiated prices are usually lower.
GPU	Price	Basis	Source
L40	$1.00	on-demand, billed per minute	Pricing page
A100	$1.60	on-demand, A100 SXM	Pricing page
RTX PRO 6000	$1.85	on-demand, RTX PRO 6000 SE	Pricing page
H100 PCIe	$2.50	on-demand, listed as “NVIDIA H100” (NVLink variant $2.60)	Pricing page
H100 SXM	$3.20	on-demand, billed per minute	Pricing page
H200	$3.99	on-demand, H200 SXM	Pricing page
B200	$6.00	on-demand, billed per minute	Pricing page
B300	$7.40	on-demand, billed per minute	Pricing page

Where we recommend them

Cheapest published H100, an alternative.
Best for B200 and GB200 access, our pick.

Worth knowing before you sign

SemiAnalysis noted that node remediation relied on manual SRE intervention (ClusterMAX 3.0); in 2.0 its managed Kubernetes deployment hung for hours.
On-demand networking is SR-IOV Ethernet; InfiniBand is for Secure Private Cloud clusters.
GB200 and GB300 NVL72 have no public price, and the terms behind “starting from” reservation prices are not stated.
The company warns that a fraudulent site (hyperstack.my) impersonates it; the real domain is hyperstack.cloud.

Go to Hyperstack's site

Sources

Work at Hyperstack? Send corrections with a source to editors@bestneoclouds.com. We fix verified errors within five business days.

Compare with

CoreWeave GPU cloud built around large InfiniBand training clusters (NASDAQ: CRWV)
Nebius Full-stack AI cloud with published on-demand GPU prices (NASDAQ: NBIS)
Crusoe AI cloud with self-serve NVIDIA and AMD GPUs on its own data center campuses
Nscale European AI cloud selling GB300 and B200 capacity (S-1 filed for NYSE: NSCL)
Together AI AI cloud for GPU clusters and model inference