India has made a significant move towards technological sovereignty by selecting Bengaluru-based start-up Sarvam to build the nation’s first indigenous large language model (LLM) under the ambitious ₹10,370-crore IndiaAI Mission. This decision was made just within two to three months after the mission’s announcement in January, showcasing an impressive sense of urgency and strategic ambition.
Sarvam’s swift selection from 67 applicants highlights the country’s commitment to establishing itself as a global AI powerhouse, especially as low-cost, open-source competitors like China’s DeepSeek emerge on the global stage. The government’s comprehensive investment—including 18,693 graphics processing units (GPUs), an AI safety institution, and applications focused on agriculture and education—signals a broad, future-ready approach. This mission aims to nurture local talent, strengthen AI infrastructure, and transition India from a predominantly service-based economy to a product-driven innovation leader.
While challenges related to global competitiveness and monetization remain, Sarvam’s journey marks a strong and promising start. It reflects how decisive action and focused ambition can open a new chapter in India’s technological advancement.
Building a Voice for India: Sarvam’s Unique Mission
Sarvam has been granted access to 4,000 GPUs for six months to create a powerful 70-billion-parameter AI model. This model is being designed with special attention to voice recognition, reasoning capabilities, and fluency across diverse Indian languages—a move that aligns perfectly with India’s multilingual diversity.
Sarvam plans to launch three versions:
Sarvam-Large for advanced enterprise applications,
Sarvam-Small for agile operations,
Sarvam-Edge for real-time, on-device processing.
By focusing on voice-driven AI using local datasets, Sarvam aims to empower farmers, rural communities, and everyday users to interact seamlessly with AI in their native languages. This vision resonates with the government’s goal to democratize access to technology and bridge the digital divide.
Challenges Ahead: Competing in a Global AI Market
Despite the encouraging momentum, Sarvam faces tough challenges, particularly in marketing and monetizing its closed-source model. Unlike open-source platforms like DeepSeek, Sarvam’s proprietary model, according to co-founders Vivek Raghavan and Pratyush Kumar, is aimed at achieving strategic autonomy and capturing enterprise markets.
While subscription models and API-based services present revenue opportunities, global AI players like OpenAI, Gemini, and DeepSeek already dominate the space. OpenAI’s CEO Sam Altman recently pointed out that maintaining profitability even with premium subscriptions is a major struggle, highlighting the financial hurdles that Sarvam could encounter.
Furthermore, historical examples such as Hike and Koo show how tough it can be for Indian tech products to compete internationally. Sarvam must overcome challenges related to high operational costs, talent acquisition, and the need for a robust go-to-market strategy. Additionally, closed-source models may face trust issues among enterprises, especially in critical sectors like healthcare and finance where data privacy and transparency are paramount.
Nevertheless, Sarvam’s culturally tuned, voice-centric AI platform offers a unique opportunity to stand out. With the right strategy and support, it has the potential to drive India’s next big leap in the global AI race.