Product
Getting Started with VESSL Cloud: Launch JupyterLab in 3 Minutes
Get started with JupyterLab on VESSL Cloud in under 3 minutes. No complex setup required — just launch and code.
VESSL Cloud has been rebuilt from the ground up to let you use high-performance GPUs instantly without complex infrastructure setup. Here are the key steps to get started.
What is VESSL Cloud?
It is a GPU cloud optimized for AI/ML workloads. From LLM fine-tuning to large-scale training and model experimentation. Rent only what you need.
Start developing AI models immediately without complicated configurations.
Getting Started with VESSL Cloud
1. Sign Up
Create an account at VESSL Cloud.
2. Create an Organization
An Organization is a unit for managing members, billing, and resources.
3. Create a Workspace
Credits are required to run a Workspace. Before running, please go to the Billing tab, register a card, and add credits.
You can create a workspace in 3 steps.
3-1. Select Resources
- Enter a Workspace Name.
- Select a Resource Spec (GPU/CPU) that fits your workload.
- You can check the estimated hourly cost on the spec card.
3-2. Deployment Settings
- Select a Container Image.
- You can use official images like PyTorch and CUDA, or your own custom images.
- Custom images must include SSHD and Python.
- Select or create an SSH Key.
- When created, the private key is downloaded, and the public key is saved to your account.
- Configure Ports.
- 8888 (Jupyter) and 22 (SSH) are open by default.
- Use Custom Ports to open additional ports if needed.
4. Attach at least one Persistent Volume.
- Temporary storage in a workspace may disappear when the workspace is stopped.
- To persist data, you must attach a Persistent volume (Workspace volume, Shared volume).
- The mount path is fixed to /root ($HOME). It cannot be changed.
3-3. Review & Deploy
- Check the Cost Summary based on the selected resources and volumes.
- Click Deploy to start.
- The workspace will not automatically terminate until your credit balance drops below $10.
4. Connect (Jupyter or SSH)
Check connection details in the Connect tab of the Workspace detail page.
Jupyter
- Click Open Jupyter or the displayed port to open it directly in your browser.
SSH (Terminal)
- Copy the SSH command from the Connect tab.
- Replace
/path/to/keywith the path to your downloaded private key. - Paste into your terminal and run.
You can also use VS Code Remote-SSH. Detailed instructions are available in the guide.
Custom Ports
If you opened Custom Ports, check the External Link/Port info in the Connect tab.
5. Cost Optimization Tips (Clean up)
Check the billing items by workspace status first.
- Pause: Use when you want to keep the environment but reduce costs.
- Storage costs still apply in the Pause state.
- Terminate: Use when you want to stop billing completely.
- To remove it from the list entirely, toggle View Terminated and then Delete.
Need Help?
- Guide Docs
- Technical Support: support@vessl.ai
- Enterprise Inquiries: Contact biz-us@vessl.ai for dedicated support.
Wayne Kim
Product Marketer
Try VESSL today
Build, train, and deploy models faster at scale with fully managed infrastructure, tools, and workflows.