Omneity Labs · Built from Morocco
AI that speaks your users’ language.
Build multimodal apps that understand text, images, video frames, and speech, then use your knowledge and tools to answer or act. Sawalni anchors a six-model stack built in Morocco.
Built in Morocco for Darija, Amazigh, Arabic, and multilingual use worldwide.
Explain what happens in this video.
Three video frames · input to Sawalniفالفيديو، واحد السيد واقف فواحد السوق، كيشوف فواحد الحيط عامر بحوايج ديال الجلد، قرون، و حوايج خرين. ف اللقطة اللولة، كيدوز يدو على شي حاجة ديال الجلد. ف اللقطة التانية، كيهز واحد القرن كبير. ف اللقطة التالتة، كيشوف مزيان ف القرن و كيتحرك شوية.
Recorded Sawalni API output · source video: IMAD EL MAADI / Pexels
A product you can use
Sawalni Batbot.
Give visitors answers from your own pages and documents. Sawalni handles the conversation and cites the source material in Khizana. You choose the languages, test the answers, and publish a page or widget.
Find the right service.
Hours, appointments, and request tracking. Ask in the language that works for you.
Choose your starting point
Build with Sawalni.
Bring the model into an app you already have, or publish an assistant from your own knowledge.
Sawalni API
Use Sawalni through an OpenAI-compatible chat API. Add multilingual, multimodal conversation and your own tools to an existing product.
model="sawalni-beta",
messages=[…]
)
Companion
An always-available assistant that stays with your tasks. Sawalni can guide the conversation and connect to actions in the apps around it.
Recorded interaction
See it work.
A real exchange with Sawalni. The person shares video and audio. Sawalni follows both and answers in conversation.
The person shares live video and speaks to Sawalni.
Sawalni responds to the scene and the spoken request together.
The exchange continues with a spoken answer.
Vision + Audio for @SawalniBot let's go!
— Omar Kamali (@OmarKamali) September 14, 2026
Research
Open work.
Wikilangs, public datasets, and published research extend this work beyond one API. They show where the language coverage comes from and let others build on it.
Explore the research →The next step
Build with us.
Explore the API or tell us what your users need to do.