A Japanese-optimized large language model developed by NII. Part of the llm-jp-3 series with 8 size variants (150M to 172B). Trained on llm-jp-corpus v3 and fine-tuned using SFT/DPO for instruction following (Instruct2/Instruct3).
Parameters
Undisclosed
Context Window
License
Proprietary
Release Date
2024-09-29
API Pricing
API pricing for this model is not yet available
Strengths
- •Highly optimized for Japanese language understanding and generation.
- •Represents an alpha version of the largest (172B) model in the series.
- •Based on the robust llm-jp-3 architecture and training pipeline.
Weaknesses
- •As an alpha release, it may have stability issues or lower performance than final versions.
- •Requires substantial computational resources for inference and deployment.
- •Performance on non-Japanese tasks may be limited compared to multilingual models.
Use Cases
- •Early access testing and evaluation of the largest llm-jp-3 model.
- •Research and development for applications requiring top-tier Japanese capability.
- •Providing feedback to improve subsequent final release versions.