Why in news
Bengaluru-based startup Sarvam AI has launched a new open-source large language model called Sarvam 1.
This model supports 11 languages, including several Indian languages, aiming to enhance multilingual capabilities in AI applications
Sarvam 1, a 2-billion-parameter model trained on a custom corpus to improve performance for Indic language speakers.
They employ innovative methods like synthetic data generation to address the scarcity of high-quality training data for Indian languages.
Sarvam AI
Bengaluru-based startup developer of large language models.
It builds a platform that enables users develop and evaluate companys GenAI apps.
It contributes to open-source models, datasets, and lead efforts for large-scale data curation.
COMMENTS