> ## Documentation Index
> Fetch the complete documentation index at: https://docs.u-kiyo.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Jupyter

> Use a private SSH tunnel for an available Jupyter server.

SSH access does not guarantee Jupyter is installed or already running. Ukiyo does not currently provide a verified one-click Jupyter setup in these docs.

## Check the rental first

Connect using [SSH](/guides/ssh). If you intend to use Jupyter, check whether its executable is installed:

```bash theme={null}
command -v jupyter
```

If it is absent, follow the [official Jupyter installation guide](https://jupyter.org/install) for your environment. Installing packages changes the rented machine; check disk space and compatibility first.

## If JupyterLab is installed

On the GPU machine, start it bound to loopback, with its normal authentication enabled:

```bash theme={null}
jupyter lab --no-browser --ip=127.0.0.1 --port=8888
```

This command assumes an appropriate user/environment. If the image requires additional configuration, follow its documented setup; do not disable authentication or expose the notebook publicly.

On your own computer, open a second terminal and forward the port using your actual SSH details:

```bash theme={null}
ssh -i /path/to/your-key.pem -p PORT -N -L 8888:127.0.0.1:8888 USERNAME@HOST
```

Keep the tunnel running and open `http://127.0.0.1:8888` locally. Use the authentication token/password shown by your Jupyter server, not your Ukiyo API token. If local port 8888 is busy, use `-L 8889:127.0.0.1:8888` and browse local port 8889.

This is a conditional Jupyter connection guide, not verification that every offer includes Jupyter. Save notebooks off-instance before [expiration](/guides/lifecycle).


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.