5 Sources
[1]
Indian AI Startup Sarvam Launches LLM Trained on 10 Indic Languages - MEDIANAMA
Disclaimer: This content generated by AI & may have errors or hallucinations. Edit before use. Read our Terms of use Sarvam AI, a Bengaluru-based artificial intelligence startup, has announced the launch of Sarvam 1, its latest open-source large language model (LLM) tailored for Indian
[2]
Sarvam AI Launches Sarvam-1, New Language Model Optimised for Indian Languages
Bengaluru-based Sarvam AI has launched a new large language model (LLM), Sarvam-1. This 2-billion-parameter model is optimised to support ten major Indian languages alongside English, including Bengali, Gujarati, Hindi, Kannada, Malayalam, Marathi, Oriya, Punjabi, Tamil, and Telugu, the official
[3]
Sarvam AI Launches Sarvam-1, Outperforms Gemma-2 and Llama-3.2
Indian AI startup Sarvam AI has launched Sarvam-1, the first LLM optimised specifically for Indian languages. Developed with 2 billion parameters, Sarvam-1 supports 10 major Indian languages -- Bengali, Gujarati, Hindi, Kannada, Malayalam, Marathi, Oriya, Punjabi, Tamil, and Telugu -- alongside
[4]
Sarvam AI Launches Indic Language Model 'Sarvam-1'
Sarvam AI also announced a partnership with Yotta Data Services for the Indic language model Sarvam AI has launched Sarvam-1, a 2 Bn parameter large language model built specifically for Indian languages. In a blogpost, the startup said that the model is optimised for 10 Indian languages,
[5]
Sarvam AI launches first LLM developed in India for local languages, built with NVIDIA AI
Created with NVIDIA NeMo software and trained on NVIDIA Hopper GPUs, Sarvam 1 model delivers efficient support for 11 languages to advance generative AI development across the nation. Sarvam AI has developed Sarvam 1, India's first home-grown large multilingual language model (LLM), built entirely
Share
Copy Link
Sarvam AI, an Indian startup, has introduced Sarvam-1, a large language model optimized for 10 Indian languages and English. This 2-billion-parameter model outperforms larger competitors and addresses key challenges in processing Indic languages.

Bengaluru-based startup Sarvam AI has unveiled Sarvam-1, a pioneering large language model (LLM) designed specifically for Indian languages
1
. This 2-billion-parameter model supports 10 major Indian languages alongside English, marking a significant advancement in natural language processing for the region2
.Sarvam-1 operates on a specialized tokenizer developed by Sarvam AI, trained on 4 trillion tokens using NVIDIA's H100 Tensor Core GPUs
1
. The model demonstrates exceptional performance, outperforming larger models like Gemma-2-2B and Llama-3.2-3B on standard benchmarks3
.Key achievements include:
4
3
3
The model's training corpus, Sarvam-2T, consists of approximately 2 trillion tokens evenly distributed across the supported languages, with Hindi making up about 20% of the data
2
. Sarvam AI employed advanced synthetic-data-generation techniques to create high-quality training datasets, addressing the lack of depth in existing web-crawled Indic language data2
.Sarvam-1 is designed to power a range of applications, including:
1
The base model is available for download on Hugging Face, allowing developers to create AI applications for Indic language users
4
.Related Stories
Sarvam AI partnered with Yotta Data Services for the model's development, utilizing Yotta's Shakti Cloud infrastructure
4
. The training process involved 1,024 GPUs over a five-day period, leveraging NVIDIA's NeMo framework5
.Sarvam-1 represents a milestone in India's AI journey, potentially positioning the country as a leader in AI innovation
5
. By addressing the technological gap faced by billions of Indic language speakers, the model could democratize access to advanced NLP capabilities across various sectors, including legal, public, finance, and others5
.As the first LLM trained entirely with data, research, and compute from India, Sarvam-1 marks the beginning of Sarvam AI's mission to build full-stack sovereign AI for the country
5
. This development aligns with the growing emphasis on localized AI solutions and could significantly impact the AI landscape in India and beyond.Summarized by
Navi
1
Science and Research

2
Policy and Regulation

3
Technology