AI Supercomputing

If you want the multi-cloud freedom for your hyper-converged infrastructure, without public-cloud lock-ins, then look no further. Start in minutes via a web browser, with self-registration. Immediately and securely access 24 tensor Petaflops (a Petaflop is the ability of a computer to do one quadrillion floating-point operations per second). Stay tuned – we are expanding to 200+ tensor PFLOPS later in 2020. Option to choose capacities at different data centers based on geographical, technical and compliance requirements.

Zeblok’s smart-sourced hyper-converged infrastructure currently stands at over 7,000 GPUs, from the NVIDIA GPU cluster sandbox residing at Stony Brook CEWIT and other strategic business partners, who will continue to enable us to expand our footprint and capacity. Enterprise clients are able to select from infrastructure catalogues to optimize cost, processing speed, networking, storage and latency. Similarly, they can combine existing on-premises, with from Zeblok's infrastructure components.

The Zeblok Frontier partnership program enables service providers and data center operators to rent out their infrastructure as a catalogue within the Zeblok AI Workstation. Yes, you can think of us as the Airbnb of Supercomputing. 

To summarize, AI Supercomputing provides the infrastructure components that AI projects require. This comprises infrastructure catalogues listed by MSPs (Managed Service providers), CSPs (Cloud Service Providers) and DCOs (Data Center Operators) across various geographic locations.

Dashboard to Select Optimal Edge Infrastructure

Users can choose their optimal mix of compute, storage, networking and connectivity, based on proximity for low latency (4G/5G enabled) and processing power (GPU-based or CPU-based) for their AI projects

Simple Registration Process

We have made it easy to register your infrastructure details (compute, storage, networking, certifications, SLA) to build your infrastructure catalogue

Keep It Private or Make It Public

MSPs, CSPs and DCOs can list their infrastructure publicly so others within the Zeblok ecosystem can view their offers and use their infrastructure for their AI projects. Enterprises can keep their infrastructure private, so their data scientists can utilize their compute capacity, while using algorithms and notebooks on the Zeblok platform.

Build Your Plans and Pricing

MSPs, CSPs and DCOs can build multiple plans with different compute capacities and price them individually. Our system offers an easy way to build your plans and pricing in the ecosystem.

Monitor Critical Parameters

The Zeblok platform provides monitoring of critical operability parameters and performance characteristics of your infrastructure



Tel: +1 (631) 223-8233

HQ Office:

1500 Stony Brook Road

Stony Brook, NY 11784


51 JFK Parkway

First Floor West

Short Hills, NJ 07078

    © 2020 Zeblok Computational, Inc.