Creation of Core Technology
Execute AI models directly on your dedicated GPU
Decide which models are allowed to run on your AI BOX
Reserve 100% of your GPU resources anytime for your own experimentation

Slice your hardware into a maximum of 8 Virtual Machines
Control GPU allocation: 0, 1, or 2 GPUs per VM (no oversubscription)
Users can connect or disconnect GPUs as needed, with owner permissions

VMs come with a minimal OS and Docker (Podman) support
Each VM is securely connected to Mycelium (P2P overlay network), ensuring efficient communication
Pre-built Docker images available for quick setup and deployment
Supports native Docker images and flights
Web gateways allow external connections to VM containers