New Delhi is gearing up for a transformative leap in artificial intelligence as the India AI Impact Summit 2026 prepares to host the launch of PARAM-2, a groundbreaking 17 billion parameter multilingual AI model under the BharatGen initiative.
This ambitious project, spearheaded by the Department of Science and Technology (DST), marks India’s push towards AI sovereignty. Unlike global giants like ChatGPT, PARAM-2 is tailored for India’s diverse linguistic landscape, supporting all 22 scheduled languages. Trained on the vast ‘Bharat Data Sagar’ datasets, it leverages a Mixture-of-Experts architecture to excel in complex multilingual tasks.
Professor Ganesh Ramakrishnan from IIT Bombay, a key figure in BharatGen, describes this not just as a model release but as a collaborative triumph involving researchers, institutions, government bodies, and industry partners. The goal? Empower India to shape its AI future independently.
What sets PARAM-2 apart is its deployment as a public digital good. Government departments, banks, hospitals, courts, and educational institutions can integrate it seamlessly, even in offline, secure environments. This approach prioritizes transparency, trust, and data sovereignty.
Rooted in the National Mission on Interdisciplinary Cyber-Physical Systems with initial funding of 235 crore rupees, BharatGen now gains momentum from the IndiaAI Mission’s 900 crore allocation. As India joins the elite league of nations building frontier AI models, PARAM-2 promises to revolutionize administration, education, and beyond, heralding a new era of inclusive technology.