Supervisely is a computer vision platform for data labeling, model workflows and custom vision applications, and this repository is its official Python SDK and ecosystem codebase. The SDK lets teams automate, customize and extend the platform instead of clicking through every task by hand. The project is released under the Apache-2.0 license, has about 545 stars on GitHub, and was actively updated in October 2026.
The Supervisely platform covers the computer vision lifecycle: importing images and videos, annotating them with a rich labeling toolbox, managing datasets and projects, and running applications from its open ecosystem of apps for tasks such as training, inference and data conversion. The Python SDK exposes those capabilities programmatically, so pipelines can move data in and out, launch apps, and run agents on your own machines to process data where it lives. Developer documentation lives on the Supervisely developer portal, and the package installs from PyPI with a single pip command. For organizations, the platform adds team workspaces and enterprise deployment options on top of the open-source SDK.
Because the SDK covers projects, datasets, annotations and application runs, teams commonly use it to build continuous pipelines in which new data is imported, labeled, quality-checked and exported for model training without manual uploads or downloads.
The Supervisely Python SDK is free and open source under the Apache-2.0 license. The Supervisely platform itself follows a freemium model, with a free tier and paid plans for larger teams and enterprise use.
Supervisely is for computer vision teams that label data and build vision pipelines at scale, and for developers who want to automate that platform through Python.