Networking
Dedicated Public IP
Every Spheron GPU instance receives a dedicated public IP address when deployed. There is no shared IP, NAT, or port forwarding. Each instance has its own routable IP for the lifetime of the deployment.
When a deployment reaches running status, the IP address and SSH connection command are shown in the instance details panel in the dashboard.
SSH Access
# Use the SSH command from the instance details panel in the dashboard
ssh ubuntu@198.51.100.42
# Or with an explicit key
ssh -i ~/.ssh/id_ed25519 ubuntu@198.51.100.42See SSH Connection for key setup and troubleshooting.
Port Access
All ports on your instance are open by default via its dedicated public IP. You can reach any service directly without any firewall configuration:
curl http://198.51.100.42:8000/v1/completionsSSH Tunneling
SSH tunneling forwards remote ports to your local machine without opening firewall rules or exposing services publicly.
Forward a remote port to your local machine:
# Forward port 8888 (Jupyter) to localhost
ssh -L 8888:localhost:8888 ubuntu@<ipAddress>
# Forward multiple ports at once
ssh -L 8888:localhost:8888 -L 8000:localhost:8000 ubuntu@<ipAddress>Then open http://localhost:8888 in your browser.
VS Code Port Forwarding
If you use VS Code Remote SSH, the Ports tab in VS Code lets you forward ports visually without typing SSH tunnel commands. This is the easiest way to forward Jupyter (8888) and TensorBoard (6006).
Additional Resources
- SSH Connection: Key setup and connection troubleshooting
- VS Code Remote: Remote development and port forwarding panel
- Security Best Practices: Secure your instances and data
- Startup Script: Cloud-init examples including firewall configuration
- API Reference: Retrieve IP address and SSH connection details programmatically