PyPI version Python uv Ruff Checked with mypy pre-commit License Apache 2.0
DocLang is the AI-native markup format for unstructured content — including documents, images, and more. It maps cleanly to LLM tokens while preserving structure, semantics, layout, and geometry in a single, unambiguous representation.
This repository is the home of the normative specification and the reference validator for DocLang. If you build with LLMs and VLMs on real-world content, this is where the standard lives.
The source of the specification is available in spec.md and exports to different formats can be found in the exports/ directory.
You can install the validator from PyPI:
pip install doclang
You can then validate a DocLang document as follows:
doclang validate -n my_document.dclg.xml
For more details, see the doclang/README.md.
If you use DocLang in academic or technical work, please cite the specification:
@misc{doclang_2026, title = {DocLang: Universal AI Document Format}, author = {{DocLang Project}}, year = {2026}, version = {main}, howpublished = {\url{https://github.com/doclang-project/doclang}}, }
To work on this repository — setup, tests, reference generation, releases — see CONTRIBUTING.md.
DocLang is developed in the open and supported by the LF AI & Data Foundation. Learn more about the project at doclang-project.
DocLang is licensed under the Apache License 2.0. See LICENSE for details.