# Academic research

LLM-jp-3.1-1.8b is a large language model developed by the National Institute of Informatics in Japan. Based on the LLM-jp-3 series, it incorporates instruction pre-training to enhance the instruction-following ability.

LLM-jp-3.1-13b-instruct4 is a large language model developed by the National Institute of Informatics in Japan. It significantly improves the instruction-following ability through instruction pre-training and supports multiple languages such as Japanese and English.

Llama 3.3 Swallow is a large language model with 70 billion parameters, built on Meta Llama 3.3, which enhances Japanese capabilities while retaining English capabilities.

VinaLLaMA is a large language model specifically optimized for Vietnamese, built on the Llama2 architecture.

A Chinese-English bilingual large language model jointly open-sourced by DeepLang Tech and Tsinghua NLP Lab, pre-trained on trillions of high-quality tokens with 8K context window processing capability

Open Cabrita 3B is an open-source large language model optimized for Portuguese, based on the LLaMA architecture, designed to narrow the performance gap between foreign language and English models.

LLaMA-30b is a large language model with 30 billion parameters, suitable for various natural language processing tasks.

LLaMA-7b is an open-source large language model released by Meta, with a parameter scale of 7 billion, suitable for natural language processing tasks.

Provides Traditional Chinese transformers models and natural language processing tools