BRAT vs Label Studio
BRAT
brat.nlplab.orgLabel Studio
labelstud.ioSide by side
| Figure | BRAT | Label Studio |
|---|---|---|
| Price from | Free | $99/mo |
| Free plan | Yes | Yes |
| Free trial | — | Yes |
| Open source | — | Yes |
| Web | Yes | Yes |
| Windows | — | Yes |
| macOS | Yes | Yes |
| Linux | Yes | Yes |
| API | — | Yes |
| Self-hosted | Yes | Yes |
| Plans published | 1 | 3 |
| Spec score · AI Data Labeling Tools | 8.0 (#9) | 8.9 (#1) |
| Purpose | brat is a web-based tool for structured text annotation, where annotations have fixed forms that computers can process. | Label Studio is an open source data labeling tool for labeling multiple data types and connecting machine learning models for predictions or active learning. |
| Support | The project directs usage questions to its mailing list and bug reports or feature requests to its GitHub issue tracker. | Community users are directed to Slack and GitHub issues, Starter Cloud includes a dedicated support portal, and Enterprise offers priority support with response time SLAs. |
Plans
Label Studio
- Community EditionFree
- Starter Cloud$99 a month
- EnterprisePriced on request
In their own words
BRAT
brat is a free, open-source tool for structured text annotation, with annotations represented in forms computers can process. It supports text spans, binary relations, n-ary associations, attributes, normalization annotations, and freeform notes. Teams can configure categories, types, and constraints for their tasks, and multiple annotators can work on the same document or collection while seeing one another’s edits nearly as they happen. Normalization can link annotations to external resources, and brat can incorporate output from automatic annotation services available as web services. Its server and client support Unicode across nearly 100 scripts. The tool checks edits against configured constraints and searches documents or collections, including keyword-in-context concordancing. Annotations export in brat stand-off format; visualizations can be saved as SVG, PNG, PDF, or EPS. The free v1.3 release is self-hosted under the MIT License. New server installations assume a UNIX-like environment; Windows users can run brat in a UNIX-like virtual machine. The experimental standalone server is not security-hardened and is not recommended for sensitive data or systems accessible from the internet.
Label Studio
Label Studio is an open-source data labeling tool for text, images, audio, video, time series, and multimodal data. It provides a configurable labeling interface, project templates, data management, and annotation import and export. Custom machine-learning backends can provide predictions, while Enterprise adds automated active-learning loops. Its integrations directory spans cloud storage, foundation-model APIs, computer vision, speech and audio, machine-learning frameworks, and infrastructure. Listed storage options include Amazon S3, Google Cloud Storage, Azure Blob Storage, Databricks, Redis, and local storage. APIs, a Python SDK, and webhooks support programmatic management and integration. Deployment options include on-premises or cloud installation using pip, Docker, Ubuntu, source, or Anaconda. Security documentation describes TLS for application-component interactions, HTTPS, and secured cookies. Enterprise is described as SOC 2 and HIPAA compliant and can be deployed in the cloud or on premises, with SSO and SCIM 2.0 support. Community Edition is free. Starter Cloud costs $99 a month, with each additional user at $49 a month; Enterprise pricing is custom.