Welcome to the exalsius documentation
exalsius enables cloud and on-premise infrastructure management for AI inference and training.
AI teams can use exalsius to:
- Deploy multi-cloud or on-prem/hybrid clusters for serving or developing AI models
- Provide self-service access to GPU resources within AI teams
- Run real-time scans of the on-demand GPU market to identify the best prices
We recommend to briefly go through the exalsius concepts of importing nodes, deploying clusters, and start workspaces, before diving into one of the guided tutorials.
Quick Start
Start your first AI development environment.
Manage Nodes
Import Nodes to your exalsius Node Pool.
Deploy Clusters
Deploy GPU clusters across multiple cloud providers or on-premise nodes.
Start Workspaces
Launch prepared workspaces, deploy models for inference, or run multi-node training.