Discussion about this post

User's avatar
Rajveer Kapoor's avatar

The Model Council feature is the one that stands out to me here. Having different AI models weigh in on the same question surfaces disagreements that a single model would flatten out, and for financial analysis or policy research where framing effects really matter, that feels less like a novelty and more like a genuine epistemic improvement. Curious whether Perplexity has published anything on how the models are selected or weighted in that mode.

No posts

Ready for more?