IndiaAI Mission: 34,381 GPUs onboarded from 14 empanelled service providers
New Delhi: India’s AI strategy aims to position the country as a global leader in artificial intelligence and as on date, 34,381 GPUs have been onboarded from 14 empanelled service providers, the government informed.
The government provides these GPUs at a subsidised cost. The average rate for these GPUs is about Rs 65 per GPU per hour. The price of H100 GPUs, widely used for foundational model training, is Rs 92 per GPU per hour which is significantly lower than the commercial hyper-scaler cloud providers, informed Union Minister Ashwini Vaishnaw in a reply to a question in the Lok Sabha.
India has a strong information technology ecosystem. It generates annual revenues of more than $250 billion and provides employment to more than 6 million people.
Global rankings such as Stanford AI rankings place India among the top countries in AI skills, capabilities, and policies to use AI. India is also the second-largest contributor to GitHub AI projects, showcasing its vibrant developer community.
The government launched IndiaAI mission in March 2024. It is a strategic initiative to establish a robust and inclusive AI ecosystem aligned with India’s development goals.
The ‘IndiaAI Application Development Initiative’ aims to develop AI applications for India specific challenges in sectors such as climate change and disaster management, healthcare, agriculture, governance, and assistive technologies for learning disabilities. As on date, 30 (thirty) applications have approved, said the minister.
In addition, sector-specific hackathons have been organised in partnership with other ministries and government institutions.
‘AIKosh’ is a unified data platform integrating datasets from government and non-government sources. The beta version launched in March 2025 currently features over 890 datasets, 208 AI models, and more than 13 development toolkits.
“These resources serve as building blocks for developers, allowing them to focus on core AI functionality instead of recreating modules,” said the minister.
The ‘IndiaAI Foundation Models’ initiative aims to develop India’s own Large Multimodal Models (LMMs) trained on Indian datasets and languages. This is to to ensure sovereign capability and global competitiveness in generative AI.
The IndiaAI Mission received more than 500 proposals. In the first phase, 4 start-ups have been selected. They include Sarvam AI, Soket AI, Gnani AI and Gan AI.
The ‘IndiaAI FutureSkills’ aims to develop AI skilled professionals in India by increasing the number of graduates, post-graduate and PhDs in AI domain. Over 200 students have received fellowships in the first year, with 26 partner institutes onboarding PhD students. It also envisions setting up Data and AI Labs in Tier 2 and Tier 3 cities across India.