Skip to content

Creating a Neptune project#

A Neptune project typically represents one machine-learning task. A project can contain runs, models, model versions, and project-level metadata.

Before you start#

Creating a project#

  1. In the Neptune app, click your workspace name in the top-left corner.
  2. Click New project.
  3. Enter the project information and set a privacy level:

    • Project name: Must consist of 3 to 50 alphanumerical characters and hyphens (-).


      The name is case-insensitive. For example, CV-Project and cv-project will be treated as the same.

    • Project key: Must consist of 1 to 10 capital letters.


      The project key can't be changed after the project is created.

      The key is used to identify runs and other objects.

    • Project privacy:

      • Private projects: Only people assigned to the project can see it.
        • To collaborate on private projects, you need a team workspace.
      • Workspace projects: People with access to your workspace can also access these projects.
      • Public projects: Freely available to view by anyone on the internet.
    • (optional) Set a project color and description.
    • To finish, click Create.

API tip

You can also create projects programmatically. For details, see Creating and deleting projects via API.

Next steps#

You can now record metadata to the project by setting the NEPTUNE_PROJECT environment variable to the full project name (workspace-name/project-name).

Example: If your workspace name is ml-team and your project name is classification:

export NEPTUNE_PROJECT="ml-team/classification"
export NEPTUNE_PROJECT="ml-team/classification"

Set permanently:

setx NEPTUNE_PROJECT "ml-team/classification"

Set only for the current session:

set NEPTUNE_PROJECT="ml-team/classification"

Alternatively, you can specify the project name when initializing Neptune:

# Create a run
run = neptune.init_run(

# Create a model
model = neptune.init_model(

# Interact with the project metadata through the API
project = neptune.init_project(


For more detailed instructions on storing the project name in an environment variable, see Setting the project name.