G42 unveils the new Indian NANDA

According to Amac News: NANDA is a 13-billion-parameter model trained on approximately 2.13 trillion tokens from language datasets, including Hindi, according to the G42 announcement.
With a name inspired by one of India's highest peaks, NANDA is the result of a collaboration between Inception (a G42 company), Mohammed Bin Zayed University of Artificial Intelligence - the world's first graduate research university dedicated to artificial intelligence and Cerebras Systems.
The model was trained on Condor Galaxy, one of the world's most powerful AI supercomputers for training and inference built by G42 and Cerebras.
The release of NANDA will be a major milestone in the AI realm for India, giving more than half a billion Indian speakers the opportunity to tap into the potential of generative AI.
India has strengthened its position as a global technology leader with transformative initiatives such as Digital India and Startup India under the leadership of Prime Minister Narendra Modi.
According to the announcement, the G42 is participating in this journey by launching NANDA in support of India's AI ambitions.
In August 2023, G42 launched JAIS, the world's first open source Arabic LLM. JAIS revolutionized Arabic Natural Language Processing (NLP) and opened access to native language generative AI capabilities to more than 400 million Arabic speakers worldwide.
With models ranging from 590 million to 70 billion parameters, JAIS set a new standard for linguistic AI that G42 is now looking to replicate for other regions where languages are still underserved.
Building on this success, NANDA extends G42's mission to empower India's scientific, academic and developer communities by accelerating the growth of the Indian-language AI ecosystem and ensuring broad access to AI across the region.