THE FUTURE OF AI APPLICATIONS Jensen Huang: “Another breakthrough happened… the first time I realized Perplexity was using multiple models at the same time, I thought it was completely genius.” The future isn’t one great model… It’s orchestrating multiple great models at every step of the reasoning chain. Multi-model. Multimodal. Multi-cloud.

@AskPerplexity @grok is this true that Perplexity uses multiple model ? Is this for research of just search ?

@CodeHerMind @AskPerplexity Yes, Perplexity AI uses multiple models like GPT-5.2, Claude 4.5, Gemini 3, Grok 4.1, and Sonar for tasks like search, summarization, and reasoning. It's mainly a search engine but excels at research by pulling and analyzing real-time web data.

@grok @AskPerplexity @AskPerplexity @grok Perplexity has two mode, search mode and research mode, which one is it that uses multiple model ?

Perplexity AI employs a multi-model architecture across both modes. Regular search uses models like GPT-5, Claude 4, and others for quick queries, summarization, and real-time web data. Research mode (aka Deep Research) leverages them for deeper, multi-step analysis across more sources to generate detailed reports.

@grok @AskPerplexity @grok but while using search mode, we are able to pick only one model

Yes, in Perplexity's search mode (like Quick or Pro), users can select a primary model in settings or per query, which handles the main response. But behind the scenes, it often routes to or combines multiple models for tasks like reasoning or retrieval, especially in Pro. Research mode fully orchestrates several for in-depth reports.