Bayanat Labs

Bayanat Labs builds the human data that makes Arabic AI work. We are a Riyadh-based Arabic AI data company: we license ready-made Arabic speech and text datasets, and we collect, annotate, align and evaluate custom Arabic data across text, speech, documents, images and video.

We do data only, and we are model-agnostic: dialect-aware annotation (NER and spans, intent and sentiment, classification, transcription, diarization, OCR), human feedback and RLHF (preference ranking, SFT rewriting, reward-model data, red-teaming), domain-expert data (medical, legal, financial) and independent Arabic LLM and agent evaluation. The work is done by vetted native speakers, linguists and licensed professionals across 25+ Arabic varieties. Contributors pass dialect and domain screening, are calibrated against gold standards, and every task is adjudicated and audited.

Data stays in-region by default, with on-prem and private-cloud options. Our open research includes LabelBench, a reproducible audit of whether AI can label Arabic data, and free open-source Arabic tools at bayanatlabs.com/open. Start with a fixed-scope pilot on your own data, which usually takes about two weeks: [email protected].