Live GPU Inventory

Dedicated GPU Clusters

Dedicated GPU clusters for AI training at scale — NVIDIA B300, B200, H200 and H100 SXM with flexible commitment terms.

All GPUsB300B200H200H100 SXM

Featured ClusterNVIDIA H200 NVLAvailable Now

H200 NVL

India

Enterprise H200 NVL bare metal nodes. Ubuntu 24.04 LTS pre-installed. RoCE v2 networking with NVLink bridge. Optional PFS and NFS storage.

GPU

8× H200 NVL

CPU

192 vCPUs

Memory

1,024 GB

Network

2× 200 Gbps

Storage

PFS / NFS

Ubuntu 24.04

✓ RoCE v2✓ NVLink bridge✓ India sovereign

Min 6 mo · 20% prepay

H200 ClusterAvailable Now📍 London, UK · 64 nodes

Request Quote

H200 — UK

Enterprise H200 Tensor Core dedicated servers in London. AMD EPYC 9654 dual CPU, 1,536GB RAM. 30TB data traffic included per month. 4U rackmount with power included.

GPU

8× NVIDIA H200 Tensor Core

CPU

Dual AMD EPYC 9654 2.4GHz

Memory

1,536 GB RAM

Storage

4× 960GB NVMe U.2

Network

2× 1GE + 2× 10GE

Uplink

1× 1,000 Mbps UTP

✓ UK sovereign✓ 30TB traffic included✓ 4U enterprise platform

Flexible commitment · Contact for pricing

B300 ClusterAvailable Now📍 Helsinki, Finland · 16 nodes

Request Quote

B300 Air-Cooled — Helsinki

HGX B300 NVL8 SMC Aircooled nodes in Helsinki. Dual Intel Xeon 9575F with 2.304TB DDR5-6400 RAM. EU-sovereign, immediately available.

GPU

HGX B300 NVL8 SMC

CPU

2× Intel Xeon 9575F

Memory

2.304 TB DDR5-6400

Storage

3.84 TB NVMe PCIe4

Network

100GbE QSFP56

Cooling

Air Cooled

✓ EU sovereign · GDPR✓ Immediate availability

Flexible · Contact for pricing

B200 ClusterAvailable Jun–Jul📍 United States · TBC nodes

Request Quote

B200 SXM — USA

HGX B200 SXM nodes with NVLink and NVSwitch connectivity. 180GB HBM3e per GPU, 1,440GB total GPU memory per node. 3.2 Tbps InfiniBand cluster fabric via 8× ConnectX-7 400G NICs. Supermicro 10U enterprise platform with redundant power.

GPU

8× NVIDIA B200 SXM

GPU Memory

180GB HBM3e · 1,440GB total

CPU

Dual Intel Xeon 6972P

Memory

1,536 GB DDR5-6400

Storage

30.72 TB Gen5 NVMe

Network

3.2 Tbps InfiniBand

✓ NVLink + NVSwitch✓ Redundant power✓ Supermicro 10U platform

Flexible commitment · Jun–Jul 2026

B200 ClusterAvailable May 8th📍 Canada · 63 nodes · 504 GPUs

Request Quote

B200 DLC — Canada

Dell PowerEdge XE9680L with Direct Liquid Cooling. 63 nodes, 504 GPUs total. Dual Intel Xeon Platinum 8570, 3TB RDIMM RAM. ConnectX-7 400GbE HPC networking.

GPU

8× HGX B200 SXM 180GB

CPU

Dual Xeon Platinum 8570

Memory

3 TB RDIMM DDR5-5600

Storage

8× 3.2TB NVMe U.2

HPC Network

ConnectX-7 400GbE OSFP

Cooling

Direct Liquid Cooling

✓ Dell PowerEdge XE9680L✓ Direct Liquid Cooling✓ 504 GPUs total

Contact for commitment terms

B300 ClusterAvailable 29th June📍 India & Indonesia · 1,000+ nodes

Request Quote

B300 Hyperscale — India & Indonesia

Dell PowerEdge XE9680L with Direct Liquid Cooling in Tier III DCs across India & Indonesia. 16 MW facility live. 1,000+ servers, 8,000+ GPUs. Stock confirmed — no lead time risk. 6–7 weeks from agreement signing.

GPU

HGX B300 8-GPU SXM6 270GB

CPU

Dual Xeon Platinum 8570

Memory

3 TB RDIMM DDR5-5600

Storage

8× 3.2TB NVMe U.2

Network

ConnectX-8 800GbE OSFP

Cooling

Direct Liquid Cooling

✓ 8,000+ GPUs · Tier III DC✓ 16 MW live facility✓ No lead time risk

3-year or 2-year commit · 6–7 weeks to deploy

B200 ClusterAvailable Mid-May📍 United States · 128 nodes · 1,024 GPUs

Request Quote

B200 Hyperscale — USA

128-node Blackwell NVL72 cluster with 1,024 B200 GPUs total. AMD CPU architecture, 800G Ethernet backend, 6PB solid-state storage. Vast Data platform optional. SOC 2 compliant with dedicated Kubernetes control plane.

GPU

1,024× NVIDIA B200

Architecture

Blackwell NVL72

Storage

6 PB solid-state

Backend

800G Ethernet

Throughput

2× 400 Gb/s

Compliance

SOC 2

✓ Vast Data platform optional✓ Kubernetes ready✓ SOC 2 compliant

12M · 25% prepay | 24M · 20% prepay

B200 ClusterAvailable Mid-May📍 United States · 128 nodes · 1,024 GPUs

Request Quote

B200 Hyperscale — USA

128-node Blackwell NVL72 cluster with 1,024 B200 GPUs. AMD CPU architecture, 800G Ethernet backend, 6PB solid-state storage. Vast Data platform optional. SOC 2 compliant with dedicated Kubernetes control plane and firewall gateway.

GPU

1,024× NVIDIA B200

Architecture

Blackwell NVL72

Storage

6 PB solid-state

Backend

800G Ethernet

Throughput

2× 400 Gb/s

Compliance

SOC 2

✓ Vast Data platform✓ Kubernetes ready✓ SOC 2 compliant

12M · 25% prepay | 24M · 20% prepay

Dedicated GPU Clusters

H200 NVL

H200 — UK

B300 Air-Cooled — Helsinki

B200 SXM — USA

B200 DLC — Canada

B300 Hyperscale — India & Indonesia

B200 Hyperscale — USA

B200 Hyperscale — USA

Didn't find the GPU you were looking for?