Gnani.ai Unveils Sovereign AI Stack Artha For Enterprises

Indian voice AI startup Gnani.ai has introduced Artha, a sovereign AI stack tailored for Indian enterprises and public institutions. The platform, unveiled by Vice President CP Radhakrishnan in New Delhi, merges two key components: Evon v3.3, a 30 billion parameter open-weight model adept in 11 Indian languages, and Plexus, Gnani.ai's agentic AI platform.

Artha enables businesses to leverage AI while maintaining control over their data and technology infrastructure. Gnani.ai's CEO Ganesh Gopalan revealed that Evon v3.3 is the company's first large language model launch, boasting 2 trillion tokens in training data with a focus on reasoning and Indic-language performance. The startup intends to expand the model family to 70 billion and 100 billion parameters, along with increasing language coverage to 22 languages.

Evon v3.3 is estimated to consume 40% fewer tokens for Indian-language workloads compared to similar models, thereby reducing compute costs. Artha was designed to tackle three primary challenges for Indian organizations adopting AI: data sovereignty, deployment costs, and effectiveness across Indian languages and real-world use cases.

Initial customer engagements show interest in Artha, with five out of 20 enterprises attending a recent meeting in Pune expressing intent to incorporate the model.