Skip to content
Varni

GPU capacity

NVIDIA B300 and B200 capacity on request, and NVIDIA H100, H200 and AMD Instinct capacity sourced against specific requirements.

Capacity is quoted in nodes where applicable. GPU configuration is confirmed per requirement. Capacity status last updated: 12 October 2026.

GPU catalogue

NVIDIA

AMD Instinct

Other

Specifications available on request.

Dedicated and reserved clusters

View AI infrastructure

Capacity can be dedicated to a single customer or reserved for an agreed term. Configuration, term and commercial model are confirmed per requirement.

What to include in a request

Share what you know. If you are still assessing options, tell us the constraints that are fixed.

  • Workload: training, inference, fine-tuning or HPC
  • Preferred GPU, or the performance you need
  • Number of GPUs or nodes
  • Region and expected start date
  • Term and commercial model
  • Whether the environment needs to be operated as well
Request GPU Capacity

Common questions

Is pricing shown on the site?

No. Pricing, configuration and availability are confirmed per requirement. Send your requirement and compute sales will respond.

What if I do not know the exact GPU model?

Select Other in the request form and describe the workload, scale and deadline. The right GPU can be identified from the requirement.

Can Varni operate the capacity for us?

Operations are scoped separately through Varni Managed Services, which is now taking enquiries.