“This facility will be accessible to all and will be operational in the coming days,” he confirmed.
India is set to develop its own generative AI model, joining the ranks of global players like OpenAI’s ChatGPT and China’s DeepSeek. The initiative, announced by Union IT Minister Ashwini Vaishnaw at the Utkarsh Odisha Conclave, will be powered by the India AI Compute Facility, which has secured 18,693 GPUs to drive the development of a Large Language Model (LLM) tailored specifically for India.
Vaishnaw stated, “We believe that there are at least six major developers who can develop AI models in the six to eight months on the outer limit, and four to six months on a more optimistic estimate.”
“A common compute facility is the most important component for creating a robust AI ecosystem,” Vaishnaw added, emphasising that researchers, startups, and academic institutions require high-end computational infrastructure to advance AI development. As part of the India AI Mission, the government has prioritized establishing a shared computing resource.
Vaishnaw revealed that the computing facility has exceeded initial expectations, securing nearly 19,000 GPUs instead of the originally planned 10,000. This includes 12,896 Nvidia H100 GPUs and 1,480 Nvidia H200 GPUs, some of the most powerful AI chips available. Of these, approximately 10,000 GPUs are ready for immediate use, while the remaining will be deployed gradually.
“This facility will be accessible to all and will be operational in the coming days,” he confirmed.
Ashwini Vaishnaw said the average discount from the market price for AI Compute services is 42 percent, and high-precision compute units, critical for foundational models, are at a 47 percent discount to market prices.
India is also launching its own foundational AI model, designed to address the country’s linguistic and cultural diversity while mitigating biases in datasets. Vaishnaw announced that proposals for AI model development would be invited starting today, with an expected development timeline of four to eight months.
“We believe that at least six major developers can build AI models within six to eight months, with some potentially delivering results in as early as four to six months,” he said. The minister also highlighted ongoing research in algorithmic efficiency, which could significantly lower the cost of model development.
Another key focus of the India AI Mission is the deployment of AI applications to address large-scale societal challenges. Vaishnaw announced that 18 AI-driven applications have been selected for the first round of funding. These projects fall under three primary themes: agriculture, climate change, and learning disabilities.
To ensure responsible AI development, India will also establish an AI safety institution. Unlike other countries where AI regulatory bodies operate within a single institution, India will adopt a hub-and-spoke model, allowing multiple institutions to collaborate and provide safety tools and frameworks.
