GPU Infrastructure

Engines of abundant intelligence.

Curiosity AI deploys the latest generation of NVIDIA and AMD GPUs across AI Factories for training workloads, inference, frontier AI modelling, generative AI and high-performance computing.

FLAGSHIP TRAINING

NVIDIA Blackwell B300 SuperPOD

Phase 1 anchor infrastructure delivering world-class FP4 precision dense compute for frontier LLM pre-training and massive cluster scaling.

Architecture: NVIDIA BlackwellMemory: 288GB HBM3e / GPUInterconnect: 3.2 Tbps Quantum-X800 InfiniBandAvailability: Q1 CY2027

ULTRA-DENSITY INFERENCE & TRAINING

NVIDIA GB300 NVL72 Liquid-Cooled

Rack-scale liquid-cooled powerhouse uniting 72 GPUs into a single unified NVLink domain for trillion-parameter inference and reasoning models.

Architecture: Grace Blackwell GB300Density: 72 GPUs / Liquid-Cooled RackFabric: 130 TB/s NVLink 5 (< 1.22 PUE)Availability: Q2 CY2027

OPEN SOVEREIGN COMPUTE

AMD Instinct MI400 Series

Open ecosystem high-memory compute clusters powered by ROCm software stack, engineered for generative AI fine-tuning and sovereign deployments.

Architecture: AMD CDNA 4Memory: Next-Gen High-Bandwidth HBMFabric: Ultra Ethernet Consortium (UEC)Availability: Phase 3 Roadmap

Workloads

Built for every stage of the AI lifecycle.

TrainingInferenceFrontier AI modellingGenerative AIHigh performance computing

Phase 1 configuration

5MW. Four clusters. One deployment.

The opening deployment is four SuperPOD clusters of NVIDIA Blackwell B300 GPUs (2,304 GPUs in total), installed in a partner AI Factory on a five-year colocation model.

SuperPOD clusters in Phase 1

GPUs per cluster

Total GPUs, Phase 1

Nodes per cluster · 8 GPU/node

Reserve long-termGPU capacity.

3–5 year GPU-as-a-Service and bare-metal reservations for neo-clouds and enterprise AI.

For customers Talk to us