Welcome

New to Strong Compute? Start here.

Introduction to Strong Compute

Our vision is to be the command and control for compute—what Kubernetes/VMware should be if they were great for AI. So we've built the Strong Compute Instant Super Computer (ISC).

Instant Super Computer

Strong Compute’s ISC is designed for efficient AI model training, packaged into the ISC Control Plane (website) and ISC Portal (terminal).

Access interactive Container environments, and launch Experiments for cluster-scale training. You bring the project, we bring the compute.

Get started

New users of Strong Compute can follow the Guided Onboarding checklist in the ISC Control Plane, or follow the Getting Started guide in these docs to get started.

Academic research grantees

Strong Compute research grantees also need to set up VPN access to the Strong Compute private cluster.

Benefits of Strong Compute

  • Multi-Cloud Backend: Large range of compute nodes with H100, A100, A6000, L40, L4, and other GPU types in 1x, 4x, and 8x GPU shapes.

  • Large cluster training: Training clusters with up to 48 GPUs.

  • Robust Checkpointing: Easy to use tools for atomic checkpointing and assisted training resumption.

  • Lightning Fast Dataset Transfer: Achieves 30GB/second (target 100GB/sec) moving containers, datasets, and training artifacts on and off training clusters.

  • Cost Control UI: Helps tightly scope and manage expenses.

Last updated