Community-tested models

AI models tested for Arabic

Does it actually work for Saudi, Najdi, Hijazi or Gulf Arabic? Benchmarks rarely say. The community tests these models on real speech and text — and reports back.

All ASR TTS Translation LLM Embedding Vision Other2 models

All languages ar en es fr multi pt zh

All dialects Egyptian Gulf Levantine MSA Najdi Saudi

Arabic support:All Full Partial None

All licenses Apache 2.0 AudarAI Community License v1.0 CC-BY-NC-4.0 Gemma Terms of Use Llama 4 Community License MIT Proprietary (API only)TII Falcon LLM License

SILMA TTS v1

SILMA AI

TTS

No evaluations yet

🇸🇦 Saudi-developedArabic: fullaren

Lightweight bilingual Arabic/English text-to-speech model, 150M parameters, built on the F5-TTS diffusion architecture and pretrained from scratch on tens of thousands of hours of speech. Supports voice cloning and, unusually for Arabic TTS, is released permissively for research and commercial use.

Hugging Face

NAMAA Saudi TTS

NAMAA Community

TTS

No evaluations yet

Arabic: fullSaudiar

Saudi-dialect Arabic text-to-speech with reference-audio voice prompting, aimed at everyday conversational speech rather than Modern Standard Arabic. Notable as the only open-weights model in this directory whose card explicitly claims Saudi dialect — no model anywhere was found claiming Najdi or Hijazi specifically. MIT-licensed.

Hugging Face