Bengaluru's Sarvam AI Confirms Plan To Bid For India’s Foundational AI Models—NDTV Profit Exclusive
Sarvam AI will submit a proposal to the Ministry of Electronics and Information Technology under the IndiaAI Mission, aiming to build India’s foundational AI models.
Bengaluru-based AI startup Sarvam will 'submit a proposal' to the Ministry of Electronics and Information Technology after the latter had called for a proposal to build India's own foundational AI models under the IndiaAI mission, Co-Founder Pratyush Kumar confirmed.
"We're grateful that the government is thinking about AI. We will participate in the call for proposals put out by the ministry," Kumar said while speaking on the sidelines of the NDTV Profit Conclave, adding that the company will also take part in the compute grant opportunities.
The comments come after IT Minister Ashwini Vaishnaw, last month, had said that at least six developers, startups, and teams are expected to start building this model over the next four to 10 months.
India's foundational AI model will be able to compete with the world's best, Vaishnaw emphasised, while assuring that data privacy will be a priority throughout the development process.
Sarvam AI is one of the first generative AI companies out of India, working to build Indian language voice bots and productivity tools. It is a full-stack company, building unique datasets, models for Indian languages speech and LLMs, and low-code authoring experiences for customer and professional agents.
"We will work closely with ministries. Sarvam will also seek additional help via the IndiaAI Mission for compute capacity. Building AI models remains a very capital-intensive task," he said.
Further, Kumar added that the emergence of DeepSeek has created a lot of energy in the ecosystem. "DeepSeek has made us question 'where are our models’. We need to think about India’s sovereign AI models, else we have the risk of being left out if we don’t step up, as has happened in the previous generations."
He also confirmed that Sarvam's family of foundational models will be ready by the end of this year. "Our model will encompass thinking, research and query answering. It will not require any supplements," he said.
