India has taken another step towards developing its domestic artificial intelligence ecosystem with the release of three indigenous AI models under the IndiaAI Mission. The models, named EKA-1, Dhīr 2B and Dhīr 5B, were unveiled in New Delhi on November 3, 2026, by the Ministry of Electronics and Information Technology (MeitY).
The models have been developed by two Indian technology companies selected under the IndiaAI Innovation Centre’s Foundation Models pillar. EKA-1 was developed by Soket Labs Technology and Research Private Limited, while Dhīr 2B and Dhīr 5B were developed by ZenteiQ AiTech Innovations.
The launch was attended by MeitY Secretary S. Krishnan and senior officials from the ministry and IndiaAI Mission, alongside representatives of the companies involved.
EKA-1: Multilingual Foundation Model
EKA-1 is a foundation model developed by Soket Labs with 26.3 billion parameters. According to the government, it was trained on approximately 7.5 trillion tokens and supports all 22 languages listed in the Eighth Schedule of the Indian Constitution, in addition to English.
The model is designed for applications involving mathematics, coding, reasoning and agentic tasks. Its development is intended to support more advanced AI applications that can address India’s linguistic and technological requirements.
Soket Labs has also highlighted potential applications in banking, financial services and cybersecurity, where specialised language and reasoning capabilities can support industry-specific requirements.
Dhīr 2B and Dhīr 5B: Lightweight Multilingual AI Assistants
ZenteiQ has introduced two open-source models, Dhīr 2B and Dhīr 5B. They support English and 10 Indian languages and are designed for secure deployment within an organisation’s own computing infrastructure.
One of their key features is the ability to operate in offline-capable environments. This can be useful for organisations that handle sensitive information or work in locations with limited internet connectivity.
The government release says these models can support document-based automation, knowledge retrieval and citizen or enterprise services. Their potential application areas include defence, government, industrial operations and education.
The company has also stated that the lightweight architecture allows deployment across different computing platforms, including GPUs and TPUs, with mobile-device deployment among the intended capabilities.
Government Support Under IndiaAI Mission
Both development teams were selected under the IndiaAI Innovation Centre pillar of the national mission. The government provided access to subsidised national AI computing infrastructure and support for associated research expenses.
The initiative forms part of India’s broader effort to develop domestic AI capabilities rather than depend entirely on technology developed elsewhere.
The IndiaAI Mission was approved by the Union Cabinet on March 7, 2024. It operates through seven pillars covering computing capacity, innovation centres, datasets, application development, skills, startup financing and safe and trusted AI.
MeitY Secretary S. Krishnan said the development of indigenous models strengthens India’s technological and strategic autonomy in artificial intelligence and creates opportunities to improve productivity across government, industry, research and education.
The government has indicated that platforms such as AIKosh can help make these resources accessible to researchers, students, innovators and the wider public. However, the announcement does not establish that every model is already available for unrestricted public use through the platform.
The release marks progress in India’s domestic AI development ecosystem. The practical impact will depend on how these models perform in real-world applications, how widely they are adopted and whether they can meet the reliability, security and performance requirements of organisations using them.
