Features
Developing high-performance and open large language models in Japan
High-performance Japanese LLMs
We aim to build general-purpose large language models that not only possess rich knowledge about Japan but also excel in English, mathematics, coding, and reasoning
Permissive licenses that allow commercial use
We adopt licenses with as few usage restrictions as possible and release the developed models on HuggingFace
Academic research and development
Led primarily by members of the Okazaki Laboratory and Yokota Laboratory at Institute of Science Tokyo, together with the National Institute of Advanced Industrial Science and Technology (AIST)
Open research and development
We share the recipes, training data, and experimental results needed to build high-performance foundation models, promoting AI research and applications
Achievements
Impact of Swallow projects spreading across the community (as of Feburary 2026)
2.3M
Model Downloads
540K
Dataset Downloads
135
Models
17
Datasets
Members
Naoaki Okazaki
Professor, Science Tokyo
Project leader, team leader of pre-training corpus, Web developer
Rio Yokota
Professor, Science Tokyo
Team leader of LLM training
Sakae Mizuki
Researcher, AIST / Science Tokyo
Team leader of instruction tuning, team leader of LLM evaluation
Kazuki Fujii
Master student, Science Tokyo
LLM training, post-training
Taishi Nakamura
Master student, Science Tokyo
LLM training, post-training, and evaluation
Youmi Ma
Assistant Professor, Science Tokyo
Post-training
Daisuke Oba
Specially Appointed Assistant Professor, Science Tokyo
Post-training
Susumu Ota
Researcher, Science Tokyo
Post-training
Koki Maeda
PhD student, Science Tokyo
LLM evaluation
Masanari Oi
Master student, Science Tokyo
LLM evaluation
Takumi Okamoto
Master student, Science Tokyo
Instruction tuning
Shigeki Ishida
Master student, Science Tokyo
LLM evaluation
Masaki Kawamura
Master student, Science Tokyo
Development of Language resources
Yukito Tajima
Master student, Science Tokyo
Development of Language resources
Taihei Shiotani
Master student, Science Tokyo
LLM evaluation
Hinari Shimada
Master student, Science Tokyo
LLM evaluation, safety
Koshiro Saito
Master student, Science Tokyo
LLM evaluation
Tatsuya Ichinose
Undergraduate student, Science Tokyo
LLM evaluation
Naoya Matsushita
Undergraduate student, Science Tokyo
LLM evaluation
Sora Miyamoto
Undergraduate student, Science Tokyo
LLM evaluation
Daisuke Nohara
Undergraduate student, Science Tokyo
Post-training
Yuta Katayama
Undergraduate student, Science Tokyo
Development of Language resources, instruction tuning
Nguyen Tien Dung
Undergraduate student, Science Tokyo
Development of Language resources
Hiroya Takamura
Research team leader, AIRC, AIST
Project manager
Past Members
Sangwhan Moon
Researcher, Science Tokyo
Post-training
Mengsay Loem
Master of TokyoTech
Expert of LLM evaluation
Shota Hirai
Master of TokyoTech
Expert of development of pre-training corpora
Taiki Iida
PhD of TokyoTech
Expert of tokenization for LLMs
Kakeru Hattori
Master student, Science Tokyo
Development of pre-training corpora, LLM evaluation