• Documentation
    • Search documentation
    karat

    +

    K

    API Reference ↗
    Data connectivity & integrationCode RepositoriesAnaconda

    Use Anaconda in Foundry

    You can use Anaconda's comprehensive Conda packages directly through Code Repositories and Code Workbooks, enabling access to enterprise-grade data science capabilities that streamline workflows and accelerate analytics and AI development.

    Anaconda access is generally available across Foundry enrollments with no license cost through June 29, 2027. After that date, workflows, builds, and pipelines using Anaconda require a separate customer-procured license. All use of Anaconda's core offerings in Foundry must comply with their Embedded End User License Agreement ↗. Palantir shares the following usage information with Anaconda:

    • The number of Foundry enrollments where Anaconda is in use.
    • The Foundry deployment type, such as on-premises or cloud.

    Contact Palantir Support with any questions about the availability or licensing of Anaconda on your enrollment.

    By accessing Conda packages in Code Repositories or Code Workbooks, you can:

    • Use Anaconda's secure enterprise-maintained package collection without leaving Foundry.
    • Build sophisticated AI, ML, and analytics applications with trusted, security-vetted libraries.
    • Leverage specialized data science tooling within Foundry's governed environment.

    Getting started

    In Code Repositories, Conda packages from Anaconda repositories are available in the Libraries side panel. Search for and select the package before choosing Add and install library.

    The Libraries panel lists the Anaconda packages available to add.

    Foundry automatically configures the necessary backing repository.

    Adding an Anaconda package imports the Anaconda backing repository.

    In Code Workbooks, navigate to the Conda environment tab in your workbook profile and select Add package to search for and add Anaconda packages.

    Anaconda packages can be added to a Code Workbook profile.

    Test the Conda environment within your workbook before saving profile changes to ensure dependencies resolve correctly. When you save changes to the Conda environment, Foundry prompts you to acknowledge that you have tested the configuration.