> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://docs.nvidia.com/ai-workbench/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.nvidia.com/ai-workbench/_mcp/server.

# Basic Quickstart

Use this guide to get a simple introduction to [NVIDIA AI Workbench](/ai-workbench/overview/introduction). For a more interactive learning experience, we recommend working through the **Tutorial Project**, available in your AI Workbench interface.

For the full list of quickstarts, see [Quickstart Guides](/ai-workbench/quickstart/overview).

In this quickstart, you will:

- [Open AI Workbench and Create a New Project](/ai-workbench/quickstart/quickstart-basic#open-ai-workbench-and-create-a-new-project) - Create a new AI Workbench project.
- [Use JupyterLab to Write Simple Code](/ai-workbench/quickstart/quickstart-basic#use-jupyterlab-to-write-simple-code) - Write and run a simple Python script in JupyterLab.
- [Use Dashboards to Track GPU, Disk Usage, and Project Environment Changes](/ai-workbench/quickstart/quickstart-basic#use-dashboards-to-track-gpu-disk-usage-and-project-environment-changes) - Monitor GPU usage, disk space, and environment changes.
- [Publish Your Project to a Git Server (Optional)](/ai-workbench/quickstart/quickstart-basic#publish-your-project-to-a-git-server-optional) - Commit and push your project to a Git repository.

## Prerequisites

Before starting, ensure that:

- NVIDIA AI Workbench is installed on your local system.

## Open AI Workbench and Create a New Project

1. Start AI Workbench

   - Double-click the AI Workbench icon on your desktop or search for it in your system's program list.

2. Select Your Location

   - On the **My Locations** page, select **Local**.

3. Create a New Project

   - On the **My Projects** page, click **New Project**.

4. Enter the following details in the **New Project** dialog:

   > - Name: hello-world
   > - Description: A Hello World example project
   > - Local Path: Accept the default path.

5. Click **Next**.

6. Choose a Base Environment

   - On the **NGC Catalog** tab, select **Python Basic**.
   - Click **Create**.

   For more details on available containers, see the [NVIDIA AI Workbench Containers](https://catalog.ngc.nvidia.com/containers?query=workbench).

7. Wait for the Project to Build

   - Track the build progress in the status bar.
   - Click **Building** or **Build Ready** to view logs.

## Use Dashboards to Track GPU, Disk Usage, and Project Environment Changes

AI Workbench dashboards provide real-time insights into your project's resource usage, helping you optimize performance and manage environments efficiently.

### Key Dashboard Metrics

**GPU Usage**

Monitor real-time GPU utilization for active project containers to optimize workloads.

**Disk Usage**

Track storage consumption across containers, volumes, and repositories.

**Project Changes**

View modifications, including added dependencies and environment updates.

AI Workbench dashboards provide real-time monitoring and control over your project environment. Use them to track system performance, manage workloads, and optimize resources.

- Monitor GPU and Disk Usage

  - Open the **Project Dashboard** to check GPU load and storage consumption.
  - Adjust workloads to optimize performance.

- Track Project Environment Changes

  - Use the **Cloned Project Dashboard** to review changes from the original repository.
  - Resolve environment warnings or unconfigured secrets as needed.

- Optimize Running Projects

  - The **Running Project Dashboard** provides real-time insights into resource usage and application status.
  - Start, stop, or modify containers directly from the dashboard for better resource allocation.

Using these dashboards ensures efficient workflow management, improves resource utilization, and enhances project scalability.

## Use JupyterLab to Write Simple Code

Follow these steps to create and run a simple Python script in JupyterLab within AI Workbench.

1. Start the Environment (Optional)
   \- Click **Environment > Start Environment** to manually start the container.
   \- If you skip this step, the environment starts automatically when you open JupyterLab.

2. Open JupyterLab
   \- Click **Open JupyterLab** at the top of the AI Workbench window.
   \- JupyterLab launches in your default browser.

3. Create and Run a Notebook
   \- Navigate to the code folder.
   \- Create a new notebook and select **Python 3 (ipykernel)** as the kernel.
   \- Enter the following code:

   > ```python
   > print("Hello World!")
   > ```

   - Click **Run** to execute the script.

4. Save the notebook as **hello-world.ipynb**.

   <img src="/ai-workbench/_fern-img/9fbea7080db6002106f51b9ee92c9024d7bdb239ed387913e8dcc6bb049aa79e.webp" alt="" width="768" />

5. Close JupyterLab

   - In JupyterLab, go to **File > Shut Down** and confirm the action.
   - Close the browser tab.
   - In AI Workbench, wait until **0 apps running** appears in the status bar.

6. Stop the Environment
   \- Click **Environment > Stop Environment** to shut down the container.

7. Add Code Files to Your Project

   You can add new code files to your project using:

   - Drag and drop into the AI Workbench interface (recommended for code files inside the code folder).
   - File browsers or terminal to manually move files into the project directory.
   - Mounts to map directories between your host system and the project container. For more information, see [Mounts](/ai-workbench/how-to/environments/runtime-configuration).

## Publish Your Project to a Git Server (Optional)

EAI Workbench projects are stored as Git repositories. When you're ready to save your changes, commit and push them to a Git server.

For more details, see [Connect to Git](/ai-workbench/how-to/integrations/github-gitlab).

1. Commit Your Changes
   \- Click **Commit** at the top of the AI Workbench window.

   - The **Create a Commit** window appears, listing all changes in your project.
   - Edit the commit message if needed, then click **Commit**.

   <img src="/ai-workbench/_fern-img/e54e5ca8b6fed30092fc7420f93759105b94420f1de33a7185b1f437cf904076.webp" alt="" width="768" />

2. Publish to a Git Server
   \- Click **Publish** at the top of the AI Workbench window.

   - The **Publish Project** window opens.
   - Edit any Git settings if needed, then click **Publish**.

   <img src="/ai-workbench/_fern-img/a301638900bad0b22087820370bdd78d2660b1668c10f4932e1863cc90fe715b.webp" alt="" width="768" />

3. Verify the Changes
   \- Open your Git repository manager (GitHub, GitLab, etc.) to confirm that your project has been updated.

## Next Steps

**Do a Quickstart**

- [Walkthrough: Customize Your Environment](/ai-workbench/quickstart/quickstart-environment)
- [Basic Quickstart (CLI)](/ai-workbench/quickstart/quickstart-cli)
- [Walkthrough: Project Dashboard for Onboarding, Monitoring and Managing Your Project](/ai-workbench/quickstart/quickstart-dashboards)
- [Advanced Walkthrough: Hybrid RAG](/ai-workbench/quickstart/quickstart-hybrid-rag)
- Explore [NVIDIA AI Workbench Example Projects](/ai-workbench/quickstart/example-projects)
- Explore the [NVIDIA Developer Forum](https://forums.developer.nvidia.com/c/ai-data-science/nvidia-ai-workbench/671)

**Learn the Concepts**

- [How to Work with AI Workbench Projects](/ai-workbench/how-to/projects/overview)
- [Add a Brev Instance](/ai-workbench/how-to/locations/add-brev)
- [Integrations and Credentials](/ai-workbench/reference/faq#integrations-and-credentials)
- [Applications](/ai-workbench/concepts/application-concept)