label-studio Releases

Label Studio 1.5.0 is a major release focused on dynamic labeling capabilities, allowing labels to be generated from ML predictions or database lookups instead of static lists. The release introduces image-based labels, taxonomy evolution (annotators can add new classes on-the-fly), and numerous UI/UX improvements including a 'no more tasks' screen, undo/redo for audio, and enhanced Data Manager columns. Note: This release includes long migrations (1-2 hours for large datasets).

Jun 16, 2022

Label Studio v1.4.1 introduces auto-created webhooks for ML backends that send events when annotations are created or updated, along with deprecating the `completions` parameter in the `fit()` method and adding a timeout to the `/train` API endpoint. The release also includes numerous improvements to labeling interfaces, export formats, and SDK functionality, while dropping Python 3.6 support due to the Pillow 9.0.0 upgrade.

Feb 12, 2022

Label Studio v1.4.0 is a major release introducing a Python SDK, customizable labeling hotkeys (including submit hotkey), the ability to create annotations from predictions with parent links, and significant infrastructure improvements including faster Docker builds, MinIO support, and Visual Studio Code Remote-Containers integration.

Nov 19, 2021

Label Studio v1.3.0 introduces Interactive ML-assisted labeling with pre-annotations as the major feature, along with improved semantic segmentation tools, delayed export options for large projects, and various UI/UX improvements to the data manager and labeling interface.

Sep 17, 2021

Label Studio v1.2.0 introduces webhooks support, enabling users to subscribe to event actions like TASKS_CREATED, ANNOTATION_CREATED, and PROJECT_UPDATED for seamless integration with machine learning pipelines.

Aug 18, 2021

Label Studio v1.1.1 is a minor release focused on cloud storage improvements, new tagging capabilities, and bug fixes. Key additions include recursive S3 bucket scanning, automatic cloud storage URL resolution, a new Repeater tag for dynamic labeling, and the ability to select specific ML model versions for predictions.

Aug 13, 2021

Label Studio v1.1.0 introduces significant image labeling enhancements including multi-tool support for various region types (rectangles, ellipses, polygons, brush masks, keypoints), YOLO export compatibility, and improved region management with visibility toggles. The release also adds data manager filtering by annotator, OCR interface improvements, and command-line user token creation for API integration.

Jun 30, 2021

Label Studio v1.0.2 is a bug fix and improvement release that adds new functionality including draft annotations, copy/edit capabilities, enhanced ML backend features, and improved export options, while fixing numerous issues with file handling, filtering, and UI elements.

May 19, 2021

Label Studio v1.0.1 is a minor release focused on collaboration improvements and bug fixes. It introduces user invitation functionality, local directory connections for accessing files, and annotation authorship tracking. Several critical bugs were fixed including issues with hotkeys, data manager loading, UTF-8 encoding on Windows, prediction score ordering, and annotation export/deletion behavior.

Apr 5, 2021

Label Studio v1.0.0 is a major rewrite that introduces multi-user support, multiple projects, and a streamlined UI. The application now uses a Django backend with database storage options (SQLite or PostgreSQL) instead of the previous Flask backend.

Mar 16, 2021

This is a patch release (v0.9.1.post1) that addresses three bug fixes in Label Studio. The release fixes issues with TextArea editing mechanics in the labeling interface, corrects labeling configuration support for NeMo (NVIDIA's neural modeling framework), and resolves a redirect issue with drag-and-drop functionality on the model prediction page.

Feb 18, 2021

This release focuses on cloud storage improvements (adding Azure and Redis support, plus S3 endpoint configuration), introduces Automated Speech Recognition with Nvidia NeMo ML backend, and includes various bug fixes for export duplicates, multi-session uploads, and non-ASCII language support.

Feb 16, 2021

This is a patch release (v0.9.0.post5) focused on bug fixes and one new API feature. The release includes improvements to file uploads for localhost, dependency updates, and a new task patch API endpoint.

Jan 22, 2021

This is a minor patch release (v0.9.0.post4) that includes a small feature enhancement for export functionality and a bug fix for prediction score naming.

Jan 19, 2021

This is a patch release (v0.9.0.post3) focused on bug fixes across multiple areas of the Label Studio platform. The release addresses issues with grid view image loading, API imports, AudioPlus functionality in Data Manager, storage error styling, and data type handling for configuration values.

Jan 16, 2021

This release focuses on major improvements to the data manager, import workflow, and cloud storage stability. The new data manager brings significant enhancements including ordering capabilities, subset selectors, tabs, and preview functionality.

Jan 12, 2021

Label Studio v0.8.2 introduces significant new labeling capabilities including Paragraphs for structured text with audio pre-listening, Taxonomy for nested classification trees, and improved undo/redo functionality. This release also adds project sharing capabilities, new UI settings like line numbers, and various ML backend enhancements.

Dec 1, 2020

This release introduces significant backend improvements including Server API refactoring, Blueprint support for better extension management, and an LS version auto-checker. It also adds a new Object detector ML backend for image recognition based on MMDetection.

Nov 20, 2020

This release introduces Time Series data labeling capabilities for classification, segmentation, forecasting, and anomaly detection, with support for importing data from CSV, TSV, and JSON files. Additional improvements include S3 presigned URL resolution, AWS S3 bucket region configuration in the UI, and drag-and-drop folder import functionality.

Oct 27, 2020

This release focuses on a major rewrite of the labeling engine for improved performance, stability, and usability. Key improvements include faster initialization, more predictable region operations, non-breaking region removal, and various UI/UX enhancements like relations visualizations, nested labeling improvements, and better task navigation.

Oct 13, 2020