About Nomeda
Arabic AI,
Built in the Open
Nomeda is a deep-tech AI lab focused on Arabic-first language models and coding orchestration.
We build the Fattah-Orch family — a suite of lightweight orchestrator models (0.6B–8B) that decompose Arabic and English coding requests into structured JSON task graphs. Trained on the first Egyptian Arabic software task decomposition dataset, they understand Egyptian Arabic, MSA, and English natively.
We also create specialized tokenizers (Nomeda-MSA-64K, Nomeda-Egyptian-16K) and open datasets (Fattah Orchestrator Dataset, Hindawi Arabic Sections) — all released openly on HuggingFace to accelerate Arabic AI research and real-world applications.
We believe the future of AI must be multilingual, context-aware, and built collaboratively — and that starts with strong foundations for underrepresented languages.
Find Us
Open Source
Explore the Fattah-Orch family (XS, S, M, L), Nomeda-MSA-64K, Nomeda-Egyptian-16K tokenizers, and the Fattah Orchestrator + Hindawi datasets — all open on HuggingFace.
Our Principles
How We Work
Arabic-Native by Design
Built from the ground up for Arabic and its dialects. Our models understand Egyptian Arabic and MSA natively — not as a translation afterthought — making them more accurate and contextually aware.
Open by Default
Model weights, tokenizer configurations, and datasets are publicly available on HuggingFace. We believe Arabic AI research progresses fastest when the community can build on shared foundations.
Practical by Design
From lightweight 0.6B models running on any CPU to 8B models for GPU inference — our ecosystem is built for real-world deployment, not just benchmarks. Task routing and model specialization keep costs low.
Get Involved
Check out our work on GitHub and HuggingFace, or reach out to collaborate.