doccano
Open-source text annotation for humans building ML datasets.
About
doccano is an open-source text annotation tool that has been a staple of NLP dataset work since 2018. You create a project, upload raw text, and start annotating: it supports text classification, sequence labeling for named entity recognition, and sequence-to-sequence tasks for summarization, with collaborative annotation, multi-language support, a RESTful API, and a mobile-friendly dark-themed interface.\n\nIt runs either from pip or Docker, uses SQLite by default with PostgreSQL as an option, and is MIT licensed with more than ten thousand GitHub stars. Human annotation remains the bottleneck for high-quality training data, and doccano is one of the most widely adopted open tools for clearing that bottleneck quickly.
What it can do
- Text classification, sequence labeling, and sequence-to-sequence annotation
- Collaborative annotation with multi-language and mobile support
- RESTful API for programmatic project and dataset management
- SQLite default with PostgreSQL support for larger teams
- MIT licensed with a large community and extensive docs
Setup
Install with pip install doccano, run doccano init and doccano webserver, or deploy the Docker Compose stack.
Prerequisites
Python 3.8+ for pip install, or Docker for the containerized setup.
Use cases
Works with
Trust score — 8/10
Scored 8/10 (verified tier, metrics-v1). Evidence: Verified via vendor site/docs/GitHub and reputable coverage, wave 100; Trending score 78/100 from researcher assessment.
Sources
Every fact above was checked against these sources. Where a claim comes from the provider and couldn't be independently verified, the text says so.