cognite-sdk

What it is and what it does

The Cognite Python SDK is a client library for accessing Cognite Data Fusion (CDF), an industrial data platform. It provides both synchronous and asynchronous interfaces to query and manipulate time series, assets, events, and other data types stored in CDF. Version 8 introduces full async/await support via the new AsyncCogniteClient, enabling non-blocking concurrent operations in notebooks and web frameworks like Streamlit, alongside a refactored synchronous client that now wraps the async implementation internally.

The SDK is tightly integrated with pandas and numpy, offering specialized endpoints like retrieve_dataframe and retrieve_arrays for efficient bulk data retrieval. It depends on authlib and msal for authentication, httpx for HTTP operations, protobuf for serialization, and standard utility libraries. Installation is straightforward with optional extras for data science workflows (pandas, numpy, geo, sympy, yaml).

Use it for:

• Fetch time series data from CDF into pandas DataFrames or numpy arrays for analysis in Jupyter notebooks.
• Build async data pipelines that concurrently retrieve multiple asset hierarchies or event streams without blocking.
• Integrate CDF data into Streamlit dashboards using async operations for responsive, non-blocking UI updates.
• Automate bulk uploads or transformations of industrial sensor data stored in Cognite Data Fusion.
• Query and filter assets, events, and metadata from CDF in Python applications or data workflows.

Worth the install?

AI-flagged interpretation of the facts on this page — verify before relying

Python SDK for interacting with Cognite Data Fusion (CDF), a cloud-based industrial data platform, with support for both synchronous and asynchronous API operations.

Yes. The SDK is actively maintained (released within 1 day), has no known vulnerabilities, low install friction, and permissive licensing. It is the official client for Cognite Data Fusion and essential if you work with CDF data. The async support in v8 is a significant addition for concurrent workloads. Install it if you need to access CDF from Python; skip it only if you do not use Cognite's platform.

Install

cognite-sdk on PyPI

pip install cognite-sdk

uv add cognite-sdk

poetry add cognite-sdk

Installing cognite-sdk

Before you install

Low friction installation with a pure-wheel distribution. Active maintenance (released 2026-08-13) and no known vulnerabilities. Eight runtime dependencies are all widely-used libraries (authlib, httpx, msal, protobuf, packaging, typing_extensions, tzdata, pip).

License in practice

Apache-2.0 (permissive) allows commercial and private use with minimal restrictions, making it suitable for most production and research contexts.

Quickstart

pip install cognite-sdk

from cognite.client import CogniteClient

client = CogniteClient()
tss = client.time_series.list()

Requires Python 3.10 or later (supports up to 3.14). Authentication credentials for Cognite Data Fusion are needed to connect.

Verify before relying

• Whether the async client is production-ready for all use cases or still has limitations compared to the sync client.
• Performance improvements from the new httpx-based HTTP client on platforms other than Windows.
• Whether optional dependencies (numpy, pandas, geopandas, shapely, sympy, PyYAML) are required for typical workflows or truly optional.

Package facts

Evidence: cognite_sdk-8.13.0-py3-none-any.whl

Tags

More Front-Ends packages