Compare answers from multiple AI models in one workspace. Let models answer independently, critique one another, and synthesize a clearer final response.