NVIDIA GPUs
Latest NVIDIA generations. Dedicated silicon, no time-sharing across tenants.
Sovereign AI
GPU pools, AUD-billed.
Dedicated NVIDIA GPUs capacity in Australian regions. On-demand or reserved, fine-tuning or training, single GPU or multi-node, all on a sovereign Australian backbone.
What it is GPUs you can plan around.
GPUs you can plan around.
Reserved when it counts.
Rapid deployment based on custom requirements, or reserve a pool by the week or month. Either way: dedicated silicon, predictable AUD pricing, zero inter-region egress.
Built for AI workloads at scale
On-demand velocity. Reserved performance.
Drop into shared GPU capacity for prototypes, reserve dedicated pools for production. Same API, same control plane, same compliance envelope.
Sovereign by default
All compute and storage in ap-syd-2 / ap-mel-1. No offshore data path.
Fine-tune or train
Single GPU for inference fine-tunes, multi-node clusters for foundational training.
Low-latency interconnect
Wire-speed RDMA-class fabric for collective ops at multi-node scale.
Why Amaze for GPU Sovereign silicon.
Sovereign silicon.
Predictable economics.
Amaze runs sovereign Australian GPU capacity on enterprise Dell hardware in Sydney and Melbourne. Latest NVIDIA generations, lower cost than offshore, AUD-billed, no offshore control plane.
Rapid deployment
Spin up a sovereign GPU pool faster than your standup is scheduled. No procurement queue.
AU onboarding
Lower cost than offshore
Predictable AUD pricing on reserved capacity. No FX surprises, no cross-border egress. Volume tiers above 10 GPUs.
100 % AU-billed
Compliance, in the open
ISO 27001 certified controls. Controls mapped to Australian regulatory frameworks.
ISO 27001 certified
Security & compliance
Compliance, in the open.
The frameworks that govern how we run, store and handle data. Audit reports available under NDA.
ISO 27001
CertifiedInformation security management system covering Amaze production infrastructure in Sydney and Melbourne.
Tier 3 Data Centres
AustralianEquinix and NEXTDC Tier 3+ facilities in Sydney and Melbourne. Carrier-neutral, 99.9% facility uptime.
Privacy Act 1988
CompliantAustralian privacy principles enforced across data handling. Notifiable Data Breach scheme aligned.
Run AI workloads right here in Australia.
100+ frontier and open-weight models, hosted in Australia.
AI Model catalogue → Meta Llama 3.1 70B
Meta Llama 3.1 405B
Mistral Mixtral 8x22B
Google Gemma 2 27B
DeepSeek DeepSeek-V3
Qwen Qwen 2.5 72B
Anthropic Claude
OpenAI GPT family
Cohere Command R+
Microsoft Phi-3 Medium
Stability Stable Diffusion 3
All vendors + 90 more →
GPU catalogue
NVIDIA GPUs, one catalogue, ample VRAM.
The NVIDIA GPUs covers fine-tunes, inference and large-context workloads, predictable memory bandwidth, ample VRAM per card, and ECC on every channel.
Mix freely across the catalogue. Your reservation grants pool-level capacity, not specific machines, so you can scale across the NVIDIA GPUs within the same contract.
Australian infrastructure built with leading global technology
Reserve sovereign GPU capacity in minutes.
Talk to an Australian solution architect about the workload, the silicon, and the contract terms.