Krutrim-2 is a 12B parameter multilingual LLM optimized for English and 22 Indic languages, built on the Mistral-NeMo architecture with a 128K token context window. It delivers best-in-class performance on Indic tasks, surpassing models 5-10x its size, and represents a significant leap in Krutrim’s mission to build AI models
The Krutrim Tokenizer, built using SentencePiece BPE, is optimized for Indian languages, English, and code, addressing challenges like rich morphology, complex scripts, and code-mixing. By employing extensive data cleanup and a balanced vocabulary, it ensures efficient tokenization for improved NLP performance across diverse Indian languages. Introduction Tokenization is a crucial
Have you ever wondered if the AI you use every day truly understands India and its diverse culture? Most AI models are trained primarily on English, Chinese, or European languages, which means they might miss the nuances of Indian languages and cultural contexts. That's where BharatBench comes in.
कृत्रिम AI Lab is India’s first dedicated AI frontier research lab, focused on developing cutting-edge, multilingual, and cost-efficient AI solutions tailored for India while contributing to global AI advancements. Through open collaboration, groundbreaking research, and frontier-scale AI models, कृत्रिम AI Lab aims to