Govt backs 20 indigenous AI models, clears 58 centres of excellence

Eleven national-level hackathons and innovation challenges have so far resulted in 62 AI prototypes.

NEW DELHI: The Centre has selected 20 indigenous foundation-model proposals for support and sanctioned over 93 lakh hours of subsidised computing capacity as part of its effort to reduce India’s dependence on foreign artificial intelligence infrastructure and technologies.In a written reply in Lok Sabha on Wednesday, minister of state for electronics and information technology Jitin Prasada said the proposals were selected from 506 applications received under the IndiaAI Mission. They include 12 large multimodal models and eight small language models, with the intellectual property remaining with the developers.Among the models already launched are Sarvam AI’s 30-billion and 105-billion parameter models and Indus, Gnani.AI’s speech-to-speech model, BharatGen’s multilingual models and Avataar AI’s video-generation model.The IndiaAI Mission, approved in March 2024 with an outlay of Rs 10,371.92 crore over five years, is aimed at creating domestic capabilities in computing, foundation models, applications, research and responsible AI.The government has empanelled 15 computing service providers and approved subsidised computing support for 237 projects, involving 93.18 lakh GPU hours. It has also issued a purchase order for a high-performance AI computing system of around 1.1 exaflops at the National Informatics Centre’s data centre in Shastri Park, Delhi. Eleven national-level hackathons and innovation challenges have so far resulted in 62 AI prototypes, of which 20 have been deployed in public-sector institutions. The government has also approved 58 AI centres of excellence across states and Union territories. Of these, work has begun on 22 centres in 13 states and UTs.Another 13 projects have been selected to work on AI safety issues such as bias, privacy, algorithm audits, explainability and machine unlearning.Prasada said India was also strengthening the wider technology ecosystem through the National Mission on Interdisciplinary Cyber-Physical Systems, which has an outlay of Rs 3,660 crore. Twenty-five technology innovation hubs have been established at academic institutions in areas including AI, robotics, cybersecurity, quantum technology and fintech.

On semiconductors, the minister said 12 manufacturing projects involving committed investments of Rs 1.64 lakh crore had been approved under the Semicon India programme, with three having started commercial production.The Union Cabinet approved Semicon 2.0 on July 15 with an outlay of Rs 1.27 lakh crore. The programme will support Indian-designed chips, semiconductor fabrication and packaging units, and the production of materials, gases and equipment required by the industry.The government acknowledged that India’s AI computing ecosystem currently depends on imported graphics processing units procured through approved service providers. It said the IndiaAI and semiconductor missions were intended to progressively build domestic capabilities across the AI supply chain.Citing Stanford’s Global AI Vibrancy Report of November 2025, the reply said India ranked third globally in AI competitiveness and was the second-largest contributor to AI projects on GitHub. Go to Source