ストックマークProprietary

stockmark-100b

Compare this model

A Japanese-specialized language model developed by Stockmark. It excels at generating natural Japanese text.

Parameters

Undisclosed

Context Window

License

Proprietary

Release Date

2024-05-13

API Pricing

API pricing for this model is not yet available

Strengths

  • Massive base model with deep understanding of Japanese language.
  • Strong foundation for fine-tuning on specialized Japanese datasets.
  • High capacity for learning complex linguistic patterns.

Weaknesses

  • Not instruction-tuned; requires significant fine-tuning for practical use.
  • Impractical for most direct applications due to size and cost.
  • Base model may generate inconsistent or unfocused text.

Use Cases

  • Research into large-scale Japanese language model capabilities.
  • Creating specialized expert models via fine-tuning on Japanese data.
  • Serving as a knowledge base for Japanese language analysis.