Overview of Japanese LLMs
A community-maintained, regularly updated overview of publicly available Japanese large language models and their benchmarks, listing each model's architecture, training data, publisher and license.
This repository is a community-maintained overview of Japanese large language models, models trained mainly on Japanese text, and of benchmarks for evaluating them. Entries are contributed by volunteers and partly drawn from papers and public resources, and the maintainers note that the list does not guarantee completeness or accuracy and may not always reflect the newest models.
For text-only and vision-and-language models, tables list details such as release year, architecture, context length, training data, the organization behind the model, and its license or terms of use; many models use permissive licenses such as MIT or Apache-2.0, but some carry non-commercial or vendor-specific licenses that are not fully open source. A separate, more readable web version of the same content is linked from the README.