URA-LLaMa
Large Language Models for Vietnamese
Hello everyone,
As a research team formed from members in Ho Chi Minh City University of Technology (HCMUT) - VNU-HCM and Stanford University, we are pleased to introduce our large language models to the community. We affectionately refer to those language models as URA-LLaMa. They are fine-tuned on Vietnamese datasets from Meta’s original LLaMa-2 model, including all three versions of 7B, 13B, and 70B.
We provide these models free of charge for research purposes. Our models come with evaluation results on 10 different tasks, covering various aspects and real-world usage scenarios. You can find information about our models at the following links:
- URA-LLaMa 7B: https://huggingface.co/ura-hcmut/ura-llama-7b
- URA-LLaMa 13B: https://huggingface.co/ura-hcmut/ura-llama-13b
- URA-LLaMa 70B: https://huggingface.co/ura-hcmut/ura-llama-70b
License and User Agreement: https://github.com/martinakaduc/ura-llama-public/blob/main/URA-LLaMa%20Model%20User%20Agreement.pdf Playground for URA-LLaMa 7B: https://huggingface.co/spaces/ura-hcmut/ura-llama-playground URA-LLaMa Evaluation Results (Actively updating): https://huggingface.co/spaces/ura-hcmut/ura-llama-evaluation\
If you want to contribute to the development of large language models for Vietnamese, please do not hesitate to contact us using the information below.
About the research group:
Website: https://www.ura.hcmut.edu.vn
Email: qttho dot hcmut dot edu dot vn
About the model licenses: nqduc at hcmut dot edu dot vn (CC sttruong at cs dot stanford dot edu; qttho at hcmut dot edu dot vn)
Thank you all.
Duc Q. Nguyen
CS Master Student
My research interests include Generative Models, Graph Representation Learning, and Probabilistic Machine Learning. My application interests include Natural Language Processing, Healthcare, and Education.