Rakuten Group's open-source Japanese language learning model, Rakuten AI 3.0, has sparked controversy due to reports that it is based on the DeepSeek V3 architecture.

CoinFeed reported on March 17th that Rakuten Group announced the launch of its Japanese-specialized large-scale model, Rakuten AI 3.0. The company claims that the model outperforms models like gpt-4o in several Japanese language benchmarks, including Japanese culture and history, and instructional compliance. It is developed based on open-source community models combined with its own bilingual data. However, users on X discovered that the project's Hugging Face page's config.json file contains entries like "model_type": "deepseek_v3". Furthermore, when answering related questions, the model's stance clearly leans towards China rather than Japan. This has sparked discussion among users about whether the model is based on DeepSeek and how much of its technology is independently developed in Japan.