Solutions

Bare metal instances

Harness the raw power of the latest NVIDIA GPUs, with straightforward contracts and transparent pricing.

  • 1 to 8 NVIDIA H100s or A100s per server

  • Ubuntu 20.04 or 22.04

  • Pre-installed NVIDIA drivers

  • NVMe storage

  • Direct SSH access

Containerized workloads

Starting at the most competitive prices, with ability to scale up to meet the needs of growing workloads.

  • BYO Docker image, or let us provide one with NVIDIA drivers pre-installed

  • 1-8 NVIDIA H100, A100, or A6000 GPUs

  • Base configuration (per GPU): 12 vCPU, 125GB RAM, & 1TB NVMe storage

  • Direct SSH access

Custom clusters

Fully maintained, state-of-the-art supercomputing clusters, built on the NVIDIA H100 HGX superpod reference architecture.

  • 32 to 1024 NVIDIA H100 80GB SXM GPUs

  • 3.2 Tbps InfiniBand networking

“NovoNimbus is committed to continually evolving to meet the needs of our customers. Our goal is to provide the best cloud GPU solutions for you, with transparent contracts and pricing that make sense right now and into the future.”

— NovoNimbus team

Discover our top-tier cloud GPU solutions