Gurukul AI Fine-Tuned LLaMA 3.1 8B on Indian NCERT Curriculum
September 1, 2026
Gurukul AI uses a curated dataset of 18,720 QA pairs based on NCERT textbooks to fine-tune LLaMA 3.1 8B for the Indian education system. The model is deployed within a RAG framework specifically designed to handle regional linguistic and syllabus-based requirements.
HOW THIS AFFECTS YOU
●
builderYou can leverage this syllabus-aligned dataset and RAG framework to build localized educational tools.
●
founderThis highlights a significant opportunity to build specialized AI tutoring for non-Westernized curricula.