ISLAMABAD: A Pakistani student studying in the United States has launched an artificial intelligence model designed exclusively for the Urdu language, a development its creator says could help bridge longstanding gaps in access to advanced digital tools for millions of speakers worldwide.

The project, called Qalb, is positioned as an Urdu-first large language model at a time when most generative AI systems are primarily trained on English and other widely used global languages. Supporters of language-specific models argue they can improve accuracy, cultural relevance and accessibility for users in education, business and public services.

Urdu is spoken by more than 230 million people globally, including in Pakistan, India and diaspora communities, but remains under-represented in advanced AI systems. Efforts to localize artificial intelligence have increasingly been seen as critical for widening participation in digital economies, particularly in developing countries.

“Qalb is now recognized as the world’s largest Large Language Model created exclusively for the Urdu language,” Taimoor Hassan, the project’s developer, was quoted this month in a report in state-run news agency APP. 

“Trained on a massive dataset of 1.97 billion tokens and benchmarked across seven-plus international evaluation frameworks, Qalb outperforms existing Urdu-focused AI models on key real-world performance indicators, setting a new standard for natural language processing in Pakistan,” Hassan said.

“This is a development model and in the next phase we would soon launch App for mobile and web so that people could use and benefit from Qalb ChatGPT.”

Hassan completed his undergraduate degree in computer science at FAAST University’s Peshawar campus and is currently studying for a master’s degree in computer science and software engineering at Auburn University in the United States. According to APP, he is a serial entrepreneur who has previously launched and exited multiple startups and has represented Pakistan at international technology forums.

“I had the opportunity to contribute in a small way to a much bigger mission for the country,” Hassan said.

“Together with my undergraduate roommates and teammates, Jawad Ahmed and Muhammad Awais, we are committed to continuously fine-tuning localized models for niche industries, which we believe can become a major breakthrough for Pakistan.”

Both collaborators are also graduates of FAAST University Peshawar Campus and are currently studying in Germany, APP reported.

The team behind Qalb said the model is intended to support local businesses, startups, educational platforms and voice-based digital services, arguing that meaningful innovation is no longer limited to large technology firms.

“Technology is no longer locked behind big budgets or big teams. With the right mindset, even a small group can build products that educate, automate, and serve millions,” Hassan told APP.