Benchmarks
Open benchmarks Kawn Lab built for Arabic tasks that had no reliable way to be measured.
Kawn Lab
Where Arabic AI begins.
The roots of Arabic AI grow here: models, data, benchmarks and training methods.
Publications
From Kawn Lab
NAKBA-NLP 2026 @ LREC 2026
Misraj AI at AR-MS NAKBA-NLP 2026: A State-of-the-Art VLM in Arabic Handwritten Text Recognition
arXiv (2025), LaTell 2026
Wasm: A Pipeline for Constructing Structured Arabic Interleaved Multimodal Corpora
arXiv (2025), LaTell 2026
Baseer: A Vision-Language Model for Arabic Document-to-Markdown OCR
arXiv (2025), LaTell 2026
Kuwain 1.5B: An Arabic SLM via Language Injection
COLING Workshops (2025)
Lahjawi: Arabic Cross-Dialect Translator
Islamic text & acoustic intelligence
Annotating Real-World Quranic Recitations for Mispronunciation Detection and Diagnosis
Islamic text & acoustic intelligence
ARISQU: A Dataset for Quotation Extraction and Attribution in Islamic Jurisprudential Issues
Inside the lab
Inside the lab. What we build, and what we share with everyone.
Datasets
Arabic corpora for pretraining, evaluation and multimodal work. Built by us. Shared with everyone.
Open Source
Everything we release in the open: models, tools and corpora. And where to find it.
Get started
Researching Arabic? Let's do it together.
For research partnerships, or access to our models and data.