Bhavish Aggarwal, the founder of Ola, is investing $230 million in Krutrim, his new AI startup focused on developing large language models (LLMs) for Indian languages.

This investment, primarily from Aggarwal's family office, underscores India's ambition to become a key player in the global AI arena, currently dominated by US and Chinese firms.

Aggarwal has also announced plans to raise an additional $1.15 billion in funding by next year, seeking external investors to contribute to Krutrim's growth.

Krutrim's announcement coincides with the open sourcing of its AI models and the unveiling of a partnership with Nvidia to build what it claims will be India's largest supercomputer.

The startup has released Krutrim-2, a 12-billion-parameter language model demonstrating promising results in processing Indian languages.

In sentiment analysis tests, Krutrim-2 achieved a score of 0.95 compared to 0.70 for competing models. It also achieved an 80% success rate in code generation tasks.

Krutrim has open-sourced several specialized models optimized for Indian languages, including systems for image processing, speech translation, and text search.

Aggarwal acknowledged that while Krutrim's models haven't yet reached global benchmarks, significant progress has been made in just one year.

He emphasized the importance of open-sourcing these models to foster collaboration within the Indian AI community and build a world-class AI ecosystem in India.

This initiative reflects India's broader strategy to establish a strong presence in the rapidly evolving field of artificial intelligence. The recent emergence of DeepSeek's R1 "reasoning" model, developed on a comparatively small budget, has highlighted the potential for innovation and disruption in the AI landscape.