Digital Mahabharata: Indian LLMs on the Global AI Map

Indian LLMs are no longer just a promise.

Sarvam AI, IndiaAI Mission, AI4Bharat and other teams are creating a local layer of artificial intelligence for a country where language, voice, and access to public services are as important as rankings on English benchmarks.

The attached HTML report has been moved to a secure React/Recharts component because the portal does not run external Chart.js scripts or CDN code in the article content.

The published version retains the nature of an infographic but organizes numbers according to sources: official PIB communications, Sarvam AI materials, AI4Bharat repositories, the Stanford HAI report, and BCG analyses.

The most important context is simple: India does not need to immediately beat OpenAI, Anthropic or DeepSeek in frontier models to build a valuable AI strategy.

Their biggest problem is local and enormous at the same time: handling multiple languages, scripts, dialects, voice channels, and administrative processes.

In this sense, Indian LLMs are less of an "Indian ChatGPT" and more of an attempt to create a national linguistic infrastructure.

The IndiaAI Mission was approved in March 2024 with a budget of over 10,300 crore rupees.