Perplexity has launched a significant upgrade to its Pro Search feature, introducing a new Reasoning Mode that transforms how users can conduct complex research and analysis.
The Feature
Pro Search now employs multi-step reasoning to break down intricate queries into manageable components, delivering more comprehensive and accurate results. The system can identify when additional computations or searches are needed and automatically activates Reasoning Mode to enhance the results.
Pro Search is now more powerful. Introducing Reasoning Mode!
— Perplexity (@perplexity_ai) October 22, 2024
Challenge your own curiosity. Ask multi-layered questions. Perplexity will adapt.
Try it yourself (sample queries in thread)👇 pic.twitter.com/NHlxA34nLd
Technical Capabilities
The upgrade integrates several powerful components:
- Advanced code execution and mathematical problem-solving through Wolfram|Alpha integration
- Improved programming capabilities for debugging and data analysis
- Complex problem-solving through multi-step reasoning

Accessibility and Limits
The feature is available to all users with different tiers:
- Free users can access 5 Pro Searches every 4 hours
- Pro subscribers ($20/month) get up to 600 Pro Searches daily
Real-World Applications
Pro Search excels in various professional contexts:
- Academic research and data analysis
- Legal research for case law identification
- Marketing trend analysis
- Code debugging and development support
Company Context
Perplexity AI, founded in 2022, has positioned this update as part of its strategy to compete with traditional search engines. The company has demonstrated its commitment to user feedback by continuously refining the feature based on user experiences and needs.
The upgrade represents a significant advancement in AI-powered research tools, particularly in handling complex, multi-layered queries that require deep analysis and comprehensive data gathering.