rinnaJapanese GPT
Japanese GPT 1B
A 1.3B-parameter Japanese GPT model despite the shorter 1B repository name.
Research release
How this checkpoint was trained
rinna trained it on Japanese C4, CC-100 and Wikipedia with a language-modeling objective. The repository provides weights under MIT.Japanese GPT 1B official model card
Use the documented parameter count
Practical guidance. The repository name is a rounded label. Use 1.3B when comparing storage or model scale, and measure generation quality on the domain you intend to adapt.
Specifications & access
| Detail | Value and conditions |
|---|---|
| Status | Research release |
| Access | Downloadable |
| Model identifiers | japanese-gpt-1b |
| Architecture | 1.3 billion parametersJapanese GPT 1B official model card |
| License | MIT.Japanese GPT 1B official model card |
Related models
- Japanese GPT-2 Small
rinna · Japanese GPT
A Japanese GPT-2 completion model with a larger decoder than the XSmall release.
Launch date not confirmedResearch releaseDownloadable - Japanese GPT-2 XSmall
rinna · Japanese GPT
A compact Japanese text-completion model trained on Japanese CC-100 and Wikipedia.
Launch date not confirmedResearch releaseDownloadable