Five frontier models share one thread and catch each other's mistakes. Suprmind...

https://front-wiki.win/index.php/Who_Judged_the_Suprmind_Study%E2%80%94Were_OpenAI_and_Anthropic_Involved%3F

Five frontier models share one thread and catch each other's mistakes. Suprmind helps teams test ideas, improve drafts, and skip tool-hopping