Models
Arabic foundation models from Kawn Lab: language, vision, speech, embedding and specialized NLP.
Kuwain
A 1.5B-parameter Arabic SLM built by language injection, the base layer for our downstream models.
Lahjawi
A family of Arabic dialect translation models covering dialect-to-dialect and dialect-to-MSA translation.
Mutarjim
Bidirectional Arabic–English translation built on Kuwain, for general and specialized domains.
Sadeed
Arabic diacritization built on Kuwain, restoring full Tashkeel for MSA and classical Arabic.
Kawn Embed Light
Lightweight general-purpose Arabic embeddings for semantic search, retrieval, clustering, and RAG.
Kawn Embed Islamic
Arabic embeddings optimized for retrieval across the Quran, Hadith, Fiqh, and Fatwa collections.
Kawn Embed Medical
Domain-specific Arabic embeddings for healthcare applications.
Nabr
Acoustic model for Quranic recitation analysis, covering ASR and Tajweed-aware mispronunciation detection.
Sawt Najd
High-quality Najdi Arabic text-to-speech for conversational AI and accessibility applications.
Islamic QA Classifier
Hierarchical classification that routes Fatwa and Fiqh questions to the correct legal category.
Islamic Quotation Extractor
Information extraction that identifies verbatim Quran, Hadith, and book quotations in classical and Fatwa texts.
Baseer
Arabic document intelligence: scans to structured Markdown
Let's talk about what you're trying to build.
Tell us the problem. We'll tell you honestly whether AI is the right answer.