Riyadh, March 23, 2025 – The Saudi Data and Artificial Intelligence Authority (SDAIA) announced that its generative Arabic Large Language Model (ALLaM), with 7 billion parameters, is now listed on the Hugging Face platform. This prominent global open-source repository for natural language processing (NLP) resources now hosts the nationally developed model, making it accessible to developers and researchers worldwide for AI applications.
SDAIA claims that ALLaM is one of the leading generative language models for the Arabic language globally.
This move aligns with the objectives of the Kingdom’s Vision 2030, under the leadership of HRH Prince Mohammed bin Salman bin Abdulaziz Al Saud, Crown Prince, Prime Minister, and SDAIA Board Chairman, who supports the authority’s efforts to position Saudi Arabia as a global technology hub for AI and digital innovation.
ALLaM is a large language model known for its accuracy and efficiency in understanding and processing Arabic. Its capabilities include text summarization, content generation, and other functionalities, making it a valuable tool for developing smart applications. The model was trained on a broad and diverse Arabic language dataset, enhancing its performance in various operational contexts.
SDAIA notes that the listing of ALLaM on Hugging Face is designed to make it more accessible to the scientific and technical community, enabling the creation of AI-powered applications.
The listing marks another step in the model's evolution. ALLaM was first integrated into IBM’s Watsonx platform for advanced Arabic language analysis and processing, then launched on Microsoft’s Azure cloud platform to offer developers and businesses access to cutting-edge cloud computing environments.
At the LEAP25 conference, Qualcomm, in collaboration with the ALLaM development team, announced the model’s availability on the Qualcomm AI Cloud. Additionally, the "ALLaM AI PC," a computer incorporating advanced AI technologies to enhance user interaction with the Arabic language, was unveiled.