MeitY's IndiaAI Mission released three indigenous foundation models in New Delhi on 3 November 2026 — EKA-1 by Soket Labs, and Dhīr 2B and Dhīr 5B by ZenteiQ AiTech Innovations.
EKA-1 is an open-source sparse hybrid model with 26.3 billion parameters, trained on about 7.5 trillion tokens, supporting all 22 scheduled Indian languages and English; it targets banking, financial services and cybersecurity.
Dhīr 2B and 5B are lightweight open-source models for English and 10 Indian languages, built for secure on-premise and offline use, including on mobile phones.
Both developers were selected under the IndiaAI Innovation Centre (Foundation Models) pillar and given subsidised national AI compute; the models are to be available through AIKosh.
| Feature | EKA-1 | Dhīr 2B / 5B |
|---|---|---|
| Developer | Soket Labs | ZenteiQ AiTech Innovations |
| Languages | 22 scheduled Indian languages + English | 10 Indian languages + English |
| Design | Open-source sparse hybrid, 26.3 bn parameters | Open-source lightweight, on-premise / offline |
| Sector focus | BFSI, cybersecurity | Defence, government, industry, education |
Build a domestic AI ecosystem — 'Making AI in India and Making AI Work for India'
Key: Approved by the Union Cabinet on 7 March 2024; outlay ₹10,371.92 crore over five years; seven pillars
Develop indigenous large foundation models
Key: Selected startups get subsidised GPU compute and research support
Repository of datasets, models and tools for Indian AI developers
Key: Distribution channel for the new models
Nodal ministry for the IndiaAI Mission
Implementing agency of the IndiaAI Mission, under Digital India Corporation
A large AI model trained on broad data that can be adapted to many tasks — translation, coding, question-answering — instead of being built for one. Most use the Transformer architecture. Parameters are its learned weights; tokens are the text pieces it trained on.
Simple Analogy: A broadly educated graduate who can be trained quickly for many jobs.
GS Paper III > Science and Technology > AI; indigenisation of technology
General Awareness > Science and Technology
General Awareness > Government schemes / technology
With the present state of development, Artificial Intelligence can effectively do which of the following? 1. Bring down electricity consumption in industrial units 2. Create meaningful short stories and songs 3. Disease diagnosis 4. Text-to-Speech Conversion 5. Wireless transmission of electrical energy Select the correct answer using the code given below:
Answer: 1, 3 and 4 only
Large general-purpose AI model adaptable to many downstream tasks
The learned weights of a model; a rough measure of its size
Running a model locally without sending data to the cloud — useful for sensitive data and low connectivity
IndiaAI's datasets and models platform