abab6.5
MiniMax's proprietary flagship model prior to its open-weight pivot
abab6.5 is a large language model from the Chinese company MiniMax, launched in April 2024 as the company’s flagship offering before it shifted toward open-weight releases with the M-series models. It runs on a mixture-of-experts architecture with a total parameter count in the trillions, though MiniMax has not disclosed exact figures, and it handles a 200,000-token context window. The lineup also included a companion model, abab6.5s, tuned for lower latency and cheaper inference at similar quality.
MiniMax built abab6.5 mainly for chat and agent products inside its own apps, including Talkie and the MiniMax Chat assistant, rather than as a developer-first API play. It arrived at a point when Chinese labs were racing to match GPT-4-class output on Chinese and English benchmarks, and MiniMax positioned it as competitive with contemporaneous frontier models on reasoning and long-form writing tasks. The model has since been superseded within MiniMax’s own catalog by newer abab7 and open-weight M-series releases, making abab6.5 mostly a historical marker of the company’s proprietary-first era.