News
EveryLLM

Indian languages

Hindi, Tamil, Bengali and the other languages of India: Sarvam, Krutrim, BharatGen, Airavata. 27 models on EveryLLM so far.

Indic models are built for the languages of India: Hindi, Bengali, Telugu, Marathi, Tamil, Urdu, Gujarati, Kannada, Odia, Malayalam, Punjabi, Assamese and others, 22 of them listed in the Eighth Schedule of the Constitution. Many Indians also mix languages, as in Hinglish, and type Indian languages in the Latin alphabet. Little of this appears in global training data, which drives the push for models trained in India.

Builders

Bengaluru's Sarvam AI released OpenHathi, a Hindi model built on Llama 2, in December 2023, and Sarvam-1, a 2-billion-parameter model for ten Indian languages and English, in October 2024. In February 2026 it showed Sarvam-30B and Sarvam-105B, mixture-of-experts models trained from scratch and then released under Apache 2.0. Ola's Krutrim launched in December 2023. AI4Bharat at IIT Madras builds open datasets and translation models for all 22 scheduled languages.

Government programmes

The IndiaAI Mission, approved in March 2024 with ₹10,371.92 crore, subsidises GPU compute and funds foundation models. It chose Sarvam in April 2025 and Soket AI, Gnani.ai and Gan.ai in May; in September 2025 the IIT Bombay-led BharatGen consortium received the largest award, ₹988.6 crore, to build models of up to a trillion parameters. BharatGen had released Param-1, a 2.9-billion-parameter Hindi and English model, in 2025.

Famous indian languages models

Still to be written: Google Gemini.

Further reading

Nemotron-4-Mini-Hindi-4B-Instr 🇺🇸 United StatesNemotron-4-Mini-Hindi-4B-Instruct A small Hindi-focused chat model from NVIDIA, tuned for Indian-context answersSmall · Open weights · 2024 MuRIL 🇮🇳 IndiaMuRIL A BERT-style encoder pre-trained on 17 Indian languages and their transliterationsEncoder · Open weights · 2021 Project Indus 🇮🇳 IndiaProject Indus Open-source language model for Hindi and dialectsGeneral · Open weights IndicBART 🇮🇳 IndiaIndicBART Sequence-to-sequence model for Indic languagesRegional · Open weights · 2021 IndicTrans2 🇮🇳 IndiaIndicTrans2 Machine translation for 22 scheduled Indian languagesDomain · Open weights · 2023 IndicBERT 🇮🇳 IndiaIndicBERT Multilingual encoder model for Indian languagesEncoder · Open weights · 2022 Tamil-LLaMA WorldTamil-LLaMA Open Tamil language model based on LLaMA 2Regional · Open weights · 2023 OpenHathi 🇮🇳 IndiaOpenHathi A 7 billion parameter language model for Hindi and English.General · Open weights Ganga-1B 🇮🇳 IndiaGanga-1B Hindi-focused autoregressive language model from IIT GandhinagarRegional · Open weights Navarasa 2.0 🇮🇳 IndiaNavarasa 2.0 Instruction-tuned model for 15 Indian languages and English.Regional · Open weights wavShuka v1 🇮🇳 IndiaShuka v1 Audio-native language model for Indic languagesMultimodal · Open weights Krutrim-2 🇮🇳 IndiaKrutrim-2 A 12-billion parameter multilingual language model focused on Indian languages and context.General · Open weights · 2024 Sarvam 2B 🇮🇳 IndiaSarvam 2B Small language model for English and ten Indian languages.Small · Open weights · 2024 Pragna-1B 🇮🇳 IndiaPragna-1B Small decoder model for Hindi, Bangla, Gujarati and EnglishSmall · Open weights

The weekly digest

Every Monday: the model of the day, new pages on EveryLLM, and the conferences and anniversaries of the week ahead.