Cognite Python SDK Generator

This is the Cognite Python SDK Generator, pygen. The purpose of this package is to help developers to work with Cognite Data Fusion's (CDF) Data Models (DM) in Python.

The core functionality is to provide a Python client that matches a data model. This enables the developer for the following benefits

Client-side validation of the data before writing it to CDF.
Autocompletion is matching the data model in the integrated developer environment (IDE). This is important as it enables:
- Discoverability of a data model through Python.
- Reduced typing errors in development.
Keeping the language domain specific for the developer. Instead of working with generic concepts such as instances, nodes and edges, the developer can work with the concepts in the data model.

Documentation

See the documentation for more information.

Installation

Without any optional dependencies

To install this package without CLI support:

pip install cognite-pygen

With optional dependencies

cli This includes CLI support such that you can run the package from the command line.

pip install cognite-pygen[cli]

If using zsh:

pip install 'cognite-pygen[cli]'

Usage

The goal of the package is to have representations of all the types in a given data model with API calls to .list(), .apply(), .delete(), and .retrieve() individuals for each type.

Creating a Python SDK from a Data Model

Given a Data Model with external id Movie in the space movies in CDF, the following command will generate a Python SDK

pygen generate --space movies \
    --external-id Movie \
    --version 1 \
    --tenant-id <tenant-id> \
    --client-id <client-id> \
    --client-secret <client-secret> \
    --cdf-cluster <cdf-cluster> \
    --cdf-project <cdf-project>

If you are not using Microsoft Entra ID (Azure AD) you need to specify the parameter --token-url, --scopes and --audience instead of --tenant-id.

Dependencies

Dependencies for the generated SDK

cognite-sdk This is the basis for all requests to the Cognite Data Fusion API.
pydantic This is used for all data classes in the generated SDK.
pandas This is used for .to_pandas() methods in the generated SDK.

Dependencies for the `pygen`

jinja2 This is used for the templating of the generated SDK.
inflect This is used for the singularization/pluralization of words in the generated SDK.
typer (Optional) This is used for the CLI of the pygen package.
black (Optional) This is used to format the code generated by the pygen package.

Changelog

Wondering about previous changes to the SDK? Take a look at the CHANGELOG.

Contributing

Want to contribute? Check out CONTRIBUTING.

Metadata

Release files for cognite-pygen 1.4.2

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for cognite-pygen 1.4.2
File	Size	Uploaded
cognite_pygen-1.4.2.tar.gz	128.7 kB	Oct 5, 2026	Details

Built distribution (wheel)

Table of built distributions (wheels) for cognite-pygen 1.4.2
File	Interpreter	ABI	Platform	Reset
cognite_pygen-1.4.2-py3-none-any.whl 167.1 kB Oct 5, 2026	Python 3	none	any	Details

Total release size: 295.9 kB

Release files / cognite_pygen-1.4.2.tar.gz

Download URL	cognite_pygen-1.4.2.tar.gz
Size	128.7 kB
Tags	Source
SHA-256 checksum How to use checksums	`89227c59f76abf03db02615e8ddd449ee80177d1999d4127ed2958a1733ae413`
BLAKE2b-256 checksum How to use checksums	`d2be4e282e349407acf3ea7d248cbb865423c8f2f06b1dc63af1d6bbbc1b6d83`
Upload date	Oct 5, 2026
Uploaded using Trusted Publishing? What is trusted publishing?	No
Uploaded via	`twine/6.2.0 CPython/3.10.22`

Release files / cognite_pygen-1.4.2-py3-none-any.whl

Download URL	cognite_pygen-1.4.2-py3-none-any.whl
Size	167.1 kB
Tags	Python 3
SHA-256 checksum How to use checksums	`45cea93cfaa9cf7147ee0d7a5c45031895f7936ee0cca71f553a6b01f4ee8055`
BLAKE2b-256 checksum How to use checksums	`3e5e395b4ae74c94e4c4a714994a56c4d3d47bfda2fd90323577a4e3302b498e`
Upload date	Oct 5, 2026
Uploaded using Trusted Publishing? What is trusted publishing?	No
Uploaded via	`twine/6.2.0 CPython/3.10.22`