HOME-EN - LLM-jp
Skip to content
LLM-jp
Establishing the baseline for Japan's LLM development.
Establishing the baseline for Japan's LLM development.
Try Demo
Release
Join Meetings
Linked Lines Move.
*The catchphrases shown above were all generated by “LLM-jp-3.1-8x13b-instruct4”.
LLM-jp is a researcher community organized by the Research and Development Center for Large Language Models at the National Institute of Informatics, and the name of a Japan-origin large language model development project.
RELEASE
Models / Corpus / Tools
Models
Fine-tuned / Pre-trained / Multimodal
Open on Hugging Face
Corpora
Pre-training Corpora / Data / Construction Scripts
Open on GitLab
Tools
Tools Evaluation & Training Utilities
Open on GitHub
All RELEASE
NEWS
2026.04.21
The 28th LLM-jp Meeting
2026.03.23
The 27th LLM-jp Meeting
2026.03.19
The website has been renewed.
2026.03.13
Announcement of the 27th LLM-jp Meeting
2026.02.24
The 26th LLM-jp Meeting
2026.02.17
Announcement on The 26th LLM-jp Meeting
All NEWS
LLM-jp is a community that brings together researchers in natural language processing and computer systems to regularly share information on large language model research and development.
We are working specifically for the following purposes:
Develop open, state-of-the-art Japanese large language models and advance related research and development
Facilitate regular exchange of knowledge on model development and recent research advances among researchers in natural language processing and related fields
Promote cross-organizational collaboration among researchers based on the sharing of data, computational resources, and other resources
Release research outputs such as models, tools, and technical materials
MORE DETAIL
Why do we develop with
LLM-jp
X-twitter
Github