A Japanese-optimized large language model developed by NII. Part of a series of 8 sizes ranging from 150M to 172B parameters. Trained on the llm-jp-corpus v3 dataset, with support for Supervised Fine-Tuning and Direct Preference Optimization via Instruct2/Instruct3. Models 13B and smaller are licensed under Apache 2.0.
Parameters
Undisclosed
Context Window
License
Proprietary
Release Date
2025-03-11
API Pricing
API pricing for this model is not yet available
Strengths
- •Sparse attention for efficiency
- •Balanced performance for its size
Weaknesses
- •Constrained context window
- •Less capable than larger models
Use Cases
- •Efficient Japanese language processing
- •Edge deployment applications
- •Research on sparse attention