We are an online community of volunteer researchers dedicated to promoting equal access to multilingual AI.

Aurora-m1

Open Source Continual Pre-training for Multilingual Language and Code

Models:

• Instruct
• Red-teamed

These are experimental research models produced for academic research. See our paper cited below.

Acknowledgement:

Training was conducted on the LUMI supercomputer, using compute resources generously provided by CSC - IT Center for Science, Finland.

Citation

If you find our project useful, we hope you would kindly star our repo and cite our work as follows:

@article{taishi2024aurora-m,
  author = {Taishi Nakamura, Mayank Mishra, Simone Tedeschi, et. al},
  title = {Aurora-M: Open Source Continual Pre-training for Multilingual Language and Code },
  year = 2025,
  publisher={In Proceedings of the 31st International Conference on Computational Linguistics},
}