Gan AI, Gnan AI, Soket AI Selected by Govt to Build India’s Next Gen LLMs

As India increases it’s focus on building large language models (LLMs), the government has selected three Indian startups—Gan AI, Gnan AI, and Soket AI—to create the next wave of LLMsunder the IndiaAI Mission.

“These teams, like Sarvam AI before them, have a clear goal—to be among the best in the world in their chosen fields,” said Ashwini Vaishnaw, Union Minister for Electronics and IT, at the “India AI – Make in India, Make AI Work for India” event held in Delhi. “The targets are bold but within reach, especially as the AI world moves towards focused, high-impact models.”

These AI models are measured by the number of “parameters” they have. The more parameters, the more powerful and capable the AI. Soket AI will build India’s first open-source AI model with 120 billion parameters, designed to work in multiple Indian languages. It will be used in important areas like healthcare, education, and defence—for example, helping doctors with diagnosis, improving online learning tools, and boosting security systems.

Gnan AI will create a 14 billion parameter Voice AI model that can understand and respond in different Indian languages, even in real-time. This will help in creating smarter voice assistants and tools for people who prefer to speak rather than type.

Gan AI is working on a 70 billion parameter model that will take text-to-speech (TTS) technology to the next level—generating computer voices that sound more natural, expressive, and human-like, even better than what global tech giants offer today.

The announcement comes shortly after Sarvam AI launched Sarvam M, a 24-billion parameter model trained on Indian languages. Alongside, the government is expanding access to AI computing power, supporting startups, and promoting India-made AI hardware like CCTV cameras.

Further, the government also launched a new international startup accelerator in partnership with Station F and HEC Paris, helping ten Indian AI startups grow on a global stage. These startups are working on everything from cybersecurity and privacy tech to geospatial analytics and surveillance tools.