Skip to main content

Documentation Index

Fetch the complete documentation index at: https://runcrate.ai/docs/llms.txt

Use this file to discover all available pages before exploring further.

Once your instance is Deployed, you can connect via SSH, set up VS Code Remote, and forward ports for services like Jupyter.

​
SSH Connection

The SSH command is available on your instance’s detail page. The general format is: 
ssh root@<instance-ip> -p <port> -i ~/.ssh/your_key
For example: 
ssh root@203.0.113.42 -p 22 -i ~/.ssh/id_rsa
Some instances use a non-standard SSH port. Always check the port shown in your instance details.

​
First Connection

On your first connection, you will be asked to confirm the server fingerprint: 
The authenticity of host '203.0.113.42 (203.0.113.42)' can't be established.
ED25519 key fingerprint is SHA256:xxxxx.
Are you sure you want to continue connecting (yes/no)?
Type yes to continue. This is normal for new instances.

​
Verify GPU Access

After connecting, confirm that your GPU is available: 
# Check GPU status
nvidia-smi

# Check CUDA version
nvcc --version

# Verify PyTorch GPU access (if Finetune or Inference template)
python3 -c "import torch; print(torch.cuda.is_available()); print(torch.cuda.get_device_name(0))"

​
Port Forwarding

To access services running on your instance from your local browser, use SSH port forwarding: 
ssh root@<instance-ip> -p <port> -i ~/.ssh/your_key -L <local-port>:localhost:<remote-port>

​
Common Ports

ServiceRemote PortLocal Forward Command
Jupyter Notebook8888-L 8888:localhost:8888
TensorBoard6006-L 6006:localhost:6006
Gradio7860-L 7860:localhost:7860
VS Code Server8080-L 8080:localhost:8080
Example — forward Jupyter to your local machine: 
ssh root@203.0.113.42 -p 22 -i ~/.ssh/id_rsa -L 8888:localhost:8888
Then open http://localhost:8888 in your browser.

​
VS Code Remote SSH

VS Code can connect directly to your instance for a full IDE experience.

​
Setup Steps

  1. Install the Remote - SSH extension in VS Code.
  2. Open the Command Palette (Cmd+Shift+P / Ctrl+Shift+P).
  3. Select Remote-SSH: Connect to Host….
  4. Enter the connection string: root@<instance-ip> -p <port>.
  5. When prompted, select the SSH key to use.
  6. VS Code will install its server component on the instance and connect.

​
SSH Config (Optional)

For quicker access, add your instance to ~/.ssh/config: 
Host runcrate-instance
    HostName 203.0.113.42
    Port 22
    User root
    IdentityFile ~/.ssh/id_rsa
Then connect with: ssh runcrate-instance or select runcrate-instance from VS Code.

​
File Transfer

Use scp to copy files to and from your instance: 
# Upload a file
scp -P <port> -i ~/.ssh/your_key ./local-file.py root@<instance-ip>:/root/

# Download a file
scp -P <port> -i ~/.ssh/your_key root@<instance-ip>:/root/results.csv ./

# Upload a directory
scp -r -P <port> -i ~/.ssh/your_key ./my-project root@<instance-ip>:/root/

​
Troubleshooting

  • Verify the instance status is Deployed on the dashboard.
  • Confirm you are using the correct IP address and port from the instance details.
  • The instance may still be starting up — wait a minute and try again.
  • Ensure your SSH key is added to your Runcrate account and was selected during deployment.
  • Verify you are using the correct key file: ssh -i ~/.ssh/correct_key ...
  • Check file permissions: chmod 600 ~/.ssh/your_key
  • Check your local network and firewall settings.
  • Verify the instance IP and port on the dashboard.
  • If the instance was recently deployed, wait 1-2 minutes for networking to initialize.