NVIDIA GPUs
Latest generation, dedicated silicon.
Sovereign AI
Multi-node GPU clusters.
Tightly coupled GPU clusters for training and large-scale inference. NVIDIA silicon on a low-latency Australian backbone, lower cost than offshore, AUD-billed.
100% Australian & Compliant by Design
ISO 27001Tier 3+ Australian data centresCarrier neutral
What it is Compute, tightly coupled.
Compute, tightly coupled.
Sovereign by default.
Multi-node GPU clusters with RDMA-class interconnect. Reserved weekly, monthly or annually, AUD-billed, sovereign by default, no shared-tenancy contention.
Cluster fabric
Wire-speed interconnect. Predictable throughput.
Reserved clusters run on enterprise Dell hardware with a low-latency Australian backbone tuned for collective ops at multi-node scale.
Built for training at scale
Reserved silicon. Sovereign envelope.
Predictable per-node performance, every job, every cluster. No burst contention. Dedicated silicon under an Australian-governed control plane.
RDMA-class fabric
Wire-speed interconnect for collective ops.
Tier 3+ AU
Equinix and NEXTDC.
Sovereign by design
Australian-controlled, Australian-billed.
Enterprise grade network performance
99.9 %
Uptime SLA
Financially backed, measured monthly.
16 +
Nodes per cluster
Volume tiering kicks in above 16 nodes.
40 Gbps
Transit per region
Carrier-grade Australian backbone.
< 15 min
P1 incident response
Australian-based on-call, 24×7.
Catalogue
Run any open-weight or frontier model, in Australia.
Eight families, dozens of sizes, one Australian API. Pick the model, call the endpoint, ship the feature.
Anthropic
OpenAI
Meta · Llama
Mistral AI
Google · Gemma
Qwen
DeepSeek
Hugging Face
Cluster compute, designed for production training.
Five things make Amaze GPU clusters different from a generic public-cloud GPU instance.
Wire-speed RDMA
Low-latency interconnect tuned for collective ops. No oversubscribed paths, predictable throughput per job, every job.
Multi-node scale
Scale from a handful of GPUs to multi-thousand-node training fabrics without rearchitecting. RDMA-class fabric end-to-end.
Hybrid by design
Pair a reserved sovereign cluster with on-prem or hyperscaler infrastructure via IP Transit and VPC. Burst into AU silicon for sensitive training, keep ancillary workloads where they already sit.
Onshore data for open-source models, end-to-end
Training data, checkpoints, fine-tuned weights and evaluation artefacts all stay in Australian regions. The end-to-end open-source training lifecycle, under Australian law.
Lower TCO and gain sovereignty
Predictable AUD pricing on reserved dedicated silicon. No FX surprises, no inter-region egress, volume tiers above 16 nodes. Lower total cost than the hyperscalers, Australian jurisdiction throughout.
How to start
Reserve a cluster in four steps.
All self-service. No tickets, no offshore approvals, no time-zone tax.
01
Talk to a solution architect
Workload sizing, topology, contract terms, all Australian.
02
Pick your silicon and footprint
NVIDIA GPUs GPU nodes, Intel Xeon CPU nodes, or both.
03
Reserve and provision
Cluster live in days, not quarters. Same API as on-demand.
04
Run your training
Foundation-model training, large-scale batch, scientific compute.
Need an outcome story?
See HPC on Amaze.
GPU clusters fit inside the broader HPC overview, sovereign supercomputing, VFX rendering, scientific compute, all on the same Australian backbone.
Australian infrastructure built with leading global technology
GPU Clusters, frequently asked questions
What's the difference between GPU Compute and GPU Clusters?
GPU Compute is on-demand single-GPU access. GPU Clusters are multi-node reservations with RDMA-class interconnect for tightly coupled training jobs.
What's the RDMA fabric speed?
Wire-speed Australian backbone tuned for collective operations. Up to 40 Gbps transit per region.
Can I mix CPU and GPU nodes in one reservation?
Yes. Hybrid silicon reservations are supported, move workloads between CPU and GPU pools without contract changes.
Is the reservation flexible?
Reserve weekly, monthly or annually. No exit penalties.
Reserve a sovereign cluster, by the week or by the year.
Talk to an Australian solution architect about the silicon, the topology, and the contract terms.