Managing packages and runtime¶

Snowflake Notebooks run inside a pre-built container environment optimized for scalable AI/ML development powered by Snowflake Container Runtime.

Python versions¶

Snowflake Notebooks support Python versions from 3.10 to 3.12. When creating a notebook service, select the Python version that best fits your workload requirements.

Pre-installed Snowflake Container Runtime packages¶

Snowflake Container Runtime version 2.2 includes approximately 100 packages and libraries that support a wide range of ML development tasks inside Snowflake.

The following sections list a curated subset of pre-installed packages (40 entries per environment) available for each Python version of Snowflake Container Runtime version 2.2.

CPU version 2.2¶

The following packages are available for each Python version of CPU version 2.2:

GPU version 2.2¶

The following packages are available for each Python version of GPU version 2.2:

Installing additional packages¶

Snowflake Notebooks provides flexible approaches for managing Python packages, ranging from fully managed runtimes to custom container images. Each option offers different tradeoffs in ease of use, governance, customization, and external connectivity. This guide compares the options and highlights the scenarios where each works best.

By default, the Snowflake Container Runtime provides a Python execution environment, preinstalled with ~100 popular DS/ML packages and 1st-party Snowflake packages such as snowflake-snowpark-python. It is optimized for Snowflake-managed CPU and GPU infrastructure and recommended for most users and general-purpose notebook workloads.

Additional ways of installing packages:

From external repositories¶

After configuring External Access Integrations (EAIs) for secure repository access, you can install packages directly from external sources such as PyPI. Users have access to a comprehensive ecosystem of packages beyond the pre-installed runtime, ensuring secure connectivity to external repos.

You can run pip install in a Python cell or in the notebook terminal.

For more information, see Set up external access for Snowflake Notebooks.

From requirements.txt¶

You can specify and install required package versions in a requirements.txt file to ensure a consistent environment setup. Install them using the following command:

!pip install -r requirements.txt

From Workspace files¶

You can download or build .whl or .py files, upload them to your workspace, and install or import them.

• Wheel files (.whl): Upload the .whl file and install it:

  Copy codeExpand code block

  !pip install file_name.whl

  Show lessSee more

  Scroll to top

  If the package contains dependencies that are not already installed, upload the complete dependency tree (either directly into Workspaces or to a stage). Alternatively, attach an EAI that allows access to a repository where the package can be downloaded (for example, PyPI).

• Python files (.py): Modules stored in your workspace can be imported directly for sharing utilities and functions across notebooks. For example:

  Copy codeExpand code block

  from my_utils import my_func

  Show lessSee more

  Scroll to top

From a Snowflake stage¶

Stages provide secure and governed package deployment by leveraging existing Snowflake data storage and governance controls for package files. Use the Snowpark session to retrieve package files from a Snowflake stage into the container environment for import and use. For example:

from snowflake.snowpark.context import get_active_session
import sys

session = get_active_session()
session.file.get("@db.schema.stage_name/math_tools.py", "/tmp")

sys.path.append("/tmp")
import math_tools

math_tools.add_one(3)

Runtime management¶

Runtime pinning¶

All notebook services are pinned to the Runtime selected at creation unless you explicitly change it by editing the service. For example, a notebook service created on Runtime 2.0 will not be automatically upgraded when new Runtime versions are released.

Runtime vulnerability scanning¶

Snowflake scans the Runtime images daily for security vulnerabilities. High or critical Common Vulnerabilities and Exposures (CVEs) are addressed by releasing new Runtime versions within 30 days of detection.

Existing notebook services can continue using Runtimes with detected CVEs. However, Runtimes with known CVEs cannot be selected when creating new notebook services.

Custom runtime¶

You can build and register your own custom container images for use with Notebooks and ML Jobs. Custom images let you include specific packages, meet compliance requirements, and ensure reproducibility across environments.

For more information, see Custom runtime images.