NVIDIA GPUs
Latest NVIDIA generations. Dedicated silicon, no time-sharing.
HPC
Run all AI and ML workloads.
Tightly coupled GPU and CPU clusters for training, scientific compute and large-scale batch. NVIDIA GPUs + Intel Xeon, on a low-latency Australian backbone.
What it is Compute, tightly coupled.
Compute, tightly coupled.
Sovereign by default.
Multi-node GPU and CPU clusters with RDMA-class interconnect. Reserved weekly, monthly or annually, AUD-billed, sovereign by default, no shared-tenancy contention.
Built for training and scientific compute
Throughput you can plan around.
Predictable per-node performance, every job, every cluster. No burst contention. Dedicated silicon under Australian-governed control plane.
High-core-count CPU
Intel Xeon for batch and tightly coupled scientific workloads.
RDMA-class fabric
Wire-speed interconnect tuned for collective ops at multi-node scale.
Tier 3+ Australian
Equinix and NEXTDC. 99.9% SLA, dual-power, carrier-neutral.
Why Amaze for clusters Reserved silicon.
Reserved silicon.
Sovereign envelope.
Amaze runs sovereign Australian cluster capacity on enterprise Dell hardware in Sydney and Melbourne. Latest NVIDIA generations, high-core-count CPU silicon, lower cost than offshore alternatives, AUD-billed.
No queue contention
Dedicated reservation means dedicated silicon. Predictable throughput per job, every job.
99.9 % uptime SLA
Lower cost than offshore
Predictable AUD pricing on reserved capacity. No FX surprises, no cross-border egress. Volume tiers above 16 nodes.
100 % AU-billed
Compliance, in the open
ISO 27001 certified controls.
ISO 27001 certified
Where clusters live
Training, batch, scientific, same backbone.
Reserved clusters fit three workload shapes. Foundation-model training on multi-node NVIDIA GPUs with collective-ops-tuned fabric. Large-scale batch on high-core-count CPU pools. Scientific compute that needs predictable throughput per cycle.
Mix silicon within the same reservation. Move a workload from CPU to GPU without renegotiating the contract.
Australian infrastructure built with leading global technology
GPU & CPU Clusters, frequently asked questions
What are Amaze GPU and CPU clusters?
Reserved, tightly coupled multi-node clusters that combine NVIDIA GPUs with high-core-count Intel Xeon CPUs on an RDMA-class, low-latency interconnect. They run on sovereign Australian infrastructure and are AUD-billed.
What workloads are the clusters built for?
Three: foundation-model training on multi-node NVIDIA GPUs, large-scale batch on high-core-count CPU pools, and scientific compute that needs predictable throughput.
How long can I reserve a cluster for?
Clusters are reserved weekly, monthly or annually. Capacity is dedicated, with no shared-tenancy contention.
Can I mix GPU and CPU within one reservation?
Yes. You can mix silicon within the same reservation and move a workload from CPU to GPU without renegotiating the contract.
Reserve a GPU or CPU cluster with Amaze Intelligence.
Talk to an Australian solution architect about the silicon, the topology, and the contract terms.