# Bilingual instruction fine-tuning

Minerva is an Italian-English bilingual large language model developed by Sapienza NLP, based on the Mistral architecture with 7 billion parameters, supporting instruction-following and dialogue tasks.

An advanced Bulgarian language model based on Gemma 2 27B developed by INSAIT, featuring exceptional understanding of Bulgarian culture and language while retaining English capabilities.

Llama-3.1-8B-Instruct-Uz is an instruction-fine-tuned large language model optimized for the Uzbek language, supporting various natural language processing tasks.

This is a Tamil instruction fine-tuned version of Google's Gemma 2B model, supporting bilingual text generation in English and Tamil.

BgGPT-7B is a model based on the Bulgarian language, which performs excellently in Bulgarian language processing and provides strong support for Bulgarian language-related application scenarios.

A 7-billion-parameter bilingual (English/Malayalam) large language model optimized based on the LLaMA-2 framework, specifically designed for instruction-following tasks