RakutenAI 3.0
Rakuten Japan
Mixture of Experts model with 671B total parameters (37B active per token). Bilingual Japanese-English, using DeepSeek-V3 architecture. Apache 2.0 licensed.
Context window128K tokens
Parameters671B (37B active)
Mixture of Experts model with 671B total parameters (37B active per token). Bilingual Japanese-English, using DeepSeek-V3 architecture. Apache 2.0 licensed.