In a landmark push for India’s sovereign artificial intelligence capacity, the Government of India has announced the selection of eight teams — including Tech Mahindra, Fractal Analytics, Avataar AI, and the IIT Bombay-led BharatGen consortium – to build indigenous large language models (LLMs) under the IndiaAI Mission.
Alongside this, the Centre will establish more than 500 Data & AI Labs nationwide, aimed at empowering students, startups, and researchers with compute resources, datasets, and hands-on training opportunities.
8 Selected Projects
The projects span multilingual, multimodal, and domain-specific LLMs designed to serve agriculture, health, governance, STEM, and Indic language needs. The selected initiatives include:
- Avataar AI – Building avatar-based multimodal LLMs (up to ~70B parameters) for Indian governance, agriculture, and education.
- BharatGen (IIT Bombay consortium) – Developing multilingual, multimodal open-source models ranging from 2B to 1 trillion parameters for agriculture, health, finance, and legal sectors.
- Fractal Analytics – Designing a 70B reasoning-focused model for structured problem solving in STEM and medicine.
- Tech Mahindra (Maker’s Lab) – Creating an efficient 8B parameter Indic LLM focusing on Hindi dialects and public-sector AI agents.
- Zenteiq – Developing “BrahmAI,” a science-driven model (8B–80B parameters) for engineering intelligence and research computing.
- GenLoop – Launching Yukti, Varta, and Kavach — small but versatile LLMs (~2B parameters) for all 22 scheduled Indian languages.
- IntelliHealth – Training a 20B parameter medical LLM for EEG signal analysis and brain-computer interface research.
- Shodh AI – Building a 7B parameter model for materials discovery and experimental science.
IndiaAI Mission: Infrastructure Boost
The IndiaAI Mission, approved with an allocation of around ₹10,300 crore, is India’s most ambitious AI programme yet. It focuses on compute access, model development, skilling, and AI innovation hubs.
The government has already rolled out tens of thousands of GPUs, offering compute credits for startups, academia, and enterprises. With the new expansion, 570+ AI/data labs will be set up, ensuring access to students in Tier-2 and Tier-3 cities.
Talent & Research Push
To nurture the next generation of AI researchers, the IndiaAI Fellowship programme has been scaled to support 13,500 scholars across undergraduate, postgraduate, and PhD levels. This fellowship will give direct access to compute clusters and datasets, reducing barriers for Indian talent.
Opportunities & Challenges
This AI mission can be a turning point for India’s AI ecosystem:
- Language diversity advantage – By training LLMs across 22 Indian languages, the models will help power inclusive AI applications for governance, education, and healthcare.
- Sectoral acceleration – Domain-specific LLMs could fast-track breakthroughs in drug discovery, materials research, and diagnostics.
- Global competitiveness – Indigenous LLMs give India a strategic edge against reliance on global players like OpenAI, Anthropic, or Google.
However, challenges remain. High-quality Indic datasets are still scarce, GPU availability is tied to global supply chains, and AI safety & governance frameworks will need to keep pace with rapid scaling.
The IndiaAI Mission is expected to showcase the first wave of working models and lab networks at the upcoming India-AI Impact Summit 2026 (Feb 19–20, 2026). For India, this is not just about technology – but about establishing sovereignty in AI innovation, ensuring that the country’s unique linguistic, social, and scientific contexts are represented in future AI systems.
Related posts:
