Jatevo
Dashboard / Decentralized inference

Decentralized inference

Dedicated contributor devices for full-device rental. Pick DGX Spark or RTX PRO 6000, then rent the assigned node and receive the private endpoint, API key, usage details, or full SSH access for your own model deployment.

Live devices
2
DGX Spark and RTX PRO 6000
Default device
RTX PRO 6000
96 GB GDDR7 ECC node
Endpoint
Private
API and SSH after assignment
Billing
Hourly
No per-token meter while active
Available model cards

Hosted now, or deploy your own model

Each active device shows the model already loaded, the runtime, access mode, and what it is best for before a customer signs in.

Start rental
Custom runtime
BYOMSSH

Deploy your own model

Rent a dedicated device, upload your own weights, install your runtime, and expose a private endpoint from the assigned node.

Access
Full SSH
Runtime
Your stack
Endpoint
Private
DedicatedFull SSHRequest access