A Japanese-optimized large language model developed by NII. Part of a family spanning 8 sizes from 150M to 172B parameters. Trained on llm-jp-corpus v3. Supports SFT/DPO via Instruct2/Instruct3. Licensed under Apache 2.0 (for 13B and smaller).
Parameters
Undisclosed
Context Window
License
Proprietary
Release Date
2025-01-27
API Pricing
API pricing for this model is not yet available
Strengths
- •Wide range of model sizes for different applications
- •Robust training on a large Japanese corpus
- •Clear licensing promotes open research and use
Weaknesses
- •The 150M model may lack capacity for complex tasks
- •Performance might lag behind larger international models
- •Requires technical expertise for fine-tuning
Use Cases
- •Embedding Japanese language understanding in edge devices
- •Serving as a baseline model for Japanese NLP research
- •Lightweight tasks like text classification or keyword extraction