abab6
Abab6
abab6 is a large language model developed by MiniMax, a Chinese AI company. Released in January 2024, it was billed as China's first trillion-parameter Mixture-of-Experts (MoE) model at full scale . In benchmark tests at the time, its results were reported to have "明显优于 ChatGPT 3.5" (clearly outperformed ChatGPT 3.5) . The model was made available through MiniMax's open platform API . MiniMax itself had raised angel funding from Yunqi Capital (云启资本) back in late 2021 .
elsewhere's corpus does not cover this subject in depth beyond these launch notices from early 2024.
AI-generated — may contain errors, please verify.
Coverage
Back to Reality! Sora Breaks Through — 4 Key Questions We're Watching | Yunqi FutureScope
Creating Virtual Worlds, Returning to Real-World Problems
MiniMax Rolls Out Self-Developed "Abab6," China's First Trillion-Parameter MoE Large Language Model, at Full Scale | Oasis Vitality
Counselor on Vitality or Advisor on Life Force or more contextually (if a title/role): **Vitality Advisor** --- *(Without additional context, this is ambiguous — it could be a diplomatic title with "参赞" as counselor, or a wellness/advisory role. The phrase appears to be a fragment rather than a full sentence.)*
MiniMax's Self-Developed "Abab6" — China's First Trillion-Parameter MoE Large Language Model — Goes Fully Live | YunXun
Large parameters solve complex problems + MoE improves computational efficiency
