GPU as a service

Unparalleled high performance architecture for the most demanding computing challenges. From AI and machine learning to graphics rendering, Hut 8’s virtual infrastructure provides access to NVidia A40 GPUs to seamlessly execute your toughest workloads.

Learn more



Process massive amounts of data faster, more efficiently, and with accurate precision.

Greater memory

Storage capacity enables the processing of large volumes of data.

Cost effective

Available as a service in the Hut 8 Cloud enabling high-demand projects to pay per use.


Take your processing power to the next level

Upgrade to the latest PCIE express

Generation 4 PCI Express doubles the bandwidth of the previous generation, improving data-transfer speeds and accelerating GPU direct memory access (DMA) transfers, providing faster input/output communication of rendered data

Accelerate complex modeling

Hut 8 GPUs feature 48 GB GDDR6 Memory, with a level of performance ideal for creatives, engineers, and data scientists alike

Faster results with greater visual accuracy

CUDA, RT and Tensor cores available on the A40 provide 2X the processing power for single-precision floating point (FP32), as well as rendering of ray-traced motion blur for faster results. Tensor cores provide up to 6X the training throughput over the previous generation to accelerate AI and data science model training without additional code

Built for the data centre

GPU A40 is compatible with a wide range of data centre infrastructure, giving customers ultimate flexibility.

What can GPUs help you with?

  • Solve large, complex computing problems
  • Run as many compute intensive applications as needed
  • Conserve costs and accelerate innovation 
  • Work on complex 3D models, large datasets, and visual simulations  

Thank you for your submission

Close this window