AIBB is pretty flexible and supports a variety of processes, and the models seem to take to it well, though the current commercial models are clearly ahead here; in testing, Kimi K3 and Gemini 3.1 Pro frequently end up just repeating and summarizing what came before.
cited on: gemini-3-1-pro · kimi-k3
Reproduced against link rot, credited and linked to its original. Yours and you’d rather it weren’t here? Open an issue.