Using Perplexity Web Research
The /perplexity command helps you get up-to-the-minute information from across the internet, powered by Perplexity’s AI research capabilities.
Basic Usage
Simply type your question or research topic after the command:
/perplexity what are the latest developments in quantum computing
/perplexity who won the champions league final yesterday
/perplexity give me the current stock price of Apple and recent news
Key Features
Real-Time News Search
Get the latest news and developments on any topic:
/perplexity what are the latest climate policy changes in the EU
Academic Research
Find and summarize recent scientific papers and discoveries:
/perplexity show me recent peer-reviewed studies on gut microbiome
Market Analysis
Get current financial data and market trends:
/perplexity analyze recent trends in electric vehicle sales in Europe
Bullet Points
Request specific formatting:
/perplexity summarize SpaceX's recent launches in bullet points
Categories
Ask for categorized information:
/perplexity break down AI safety initiatives into industry, academic, and government categories
Get chronological breakdowns:
/perplexity create a timeline of fusion energy breakthroughs in 2024
Advanced Features
Source-Specific Research
Direct your research to specific websites or exclude unwanted sources:
/perplexity find news about climate change from sites: nytimes.com, reuters.com
/perplexity research quantum computing exclude: pinterest.com, quora.com
Get information from a specific time period:
/perplexity what happened in renewable energy this week
/perplexity show me AI breakthroughs from the past month
Request information in a specific structured format:
/perplexity research Hilton Hotels recent developments and return as structured JSON with categories for recent news, business performance, and new properties
File Analysis
Include files in your research (PDFs, Word documents, Excel spreadsheets):
/perplexity analyze this market report and find similar recent trends
[attach your file]
AI Model Options
You can specify which AI model to use for different research needs:
Quick Factual Queries
/perplexity using sonar: who won the NBA championship last year
Comprehensive Research
/perplexity using sonar-pro: analyze the impact of climate change on global agriculture
Complex Reasoning Tasks
/perplexity using sonar-reasoning: evaluate the pros and cons of different renewable energy sources
Customization Options
You can customize your research by including these options in natural language:
Adjust Output Length
/perplexity give me a detailed explanation of quantum computing with at least 1000 words
Control Response Style
/perplexity provide a creative explanation of blockchain technology with high creativity level
Search Context Size
/perplexity using high search context: give an in-depth analysis of machine learning trends
Domain Filtering
/perplexity research recent space missions from sites: nasa.gov, space.com, esa.int
Time Filtering
/perplexity what happened in AI research in the past week
/perplexity show me climate policy changes from the past month
Structured Data Responses
You can request structured data responses for complex information needs. This is helpful for getting organized information about a topic:
/perplexity what are the latest news and developments about Hilton Hotels? Include information about their recent expansions, business performance, and any notable changes in their services or loyalty program. Return as structured data with categories for recent developments, business performance, new properties, and loyalty program updates.
This will return a structured response with clearly organized information in categories.
Example JSON Response Structure
For developers and advanced users who need structured data, you can request specific JSON formats:
/perplexity research recent AI developments and output in JSON format with categories for breakthroughs, companies, and applications
The system will return a structured JSON response like:
{
"breakthroughs": [
{
"title": "New Large Language Model",
"description": "Details about the breakthrough",
"date": "2023-05-15"
}
],
"companies": [
{
"name": "AI Corp",
"focus": "Natural language processing",
"recent_developments": "Description of recent work"
}
],
"applications": {
"healthcare": "Description of healthcare applications",
"finance": "Description of finance applications"
}
}
Advanced JSON Schema Example
For complex research needs, you can specify a detailed JSON schema structure. Here’s an example for researching hotel company information:
/perplexity research Hilton Hotels with detailed JSON schema for recent developments, business performance, new properties, and loyalty program updates
Prodide your own schema like this:
{
"type": "json_schema",
"json_schema": {
"schema": {
"type": "object",
"properties": {
"recent_developments": {
"type": "array",
"items": {
"type": "object",
"properties": {
"title": { "type": "string" },
"summary": { "type": "string" },
"date": { "type": "string" }
}
}
},
"business_performance": {
"type": "object",
"properties": {
"recent_revenue": { "type": "string" },
"stock_performance": { "type": "string" },
"growth_outlook": { "type": "string" }
}
},
"new_properties": {
"type": "array",
"items": {
"type": "object",
"properties": {
"location": { "type": "string" },
"opening_date": { "type": "string" },
"property_type": { "type": "string" }
}
}
},
"loyalty_program_updates": { "type": "string" }
},
"required": ["recent_developments", "business_performance"]
}
}
}
This will generate a highly structured response that matches exactly the format you’ve specified, making it ideal for data analysis or integration with other systems.
Best Practices
-
Be Specific: Instead of “tell me about AI”, try “what are the latest developments in AI language models for medical diagnosis”
-
Request Format: Specify how you want the information presented:
/perplexity explain quantum computing breakthroughs
format as bullet points
focus on practical applications
-
Time-Sensitive Queries: For news and current events, specify timeframes:
/perplexity what happened in renewable energy this week
-
Multi-Part Questions: Break complex queries into connected parts:
/perplexity analyze the chip manufacturing industry
focus on: new facilities, major players, investment trends
timeline: past 6 months
-
Request Specific Sources: Direct research to trusted domains:
/perplexity research climate change from sites: science.org, nature.com
Tips for Quality Results
- Use clear, specific questions
- Specify your preferred format (bullet points, categories, timeline, JSON)
- Indicate if you want academic, news, or business focus
- Mention time periods when relevant (day, week, month, year)
- Request specific types of sources if needed
- Ask for comparisons or trends rather than just facts
- Specify the depth of analysis you want
- Request a specific AI model for your particular needs
Limitations
- Cannot access private or paywalled content
- May have a slight delay in real-time events
- Best for factual research rather than opinion-based topics
- Sources are limited to publicly available information
- File processing only works with documents (PDF, Word, Excel, PowerPoint, CSV)
- Image processing is not supported
Example Use Cases
Scientific Research
/perplexity what are the breakthrough discoveries in CRISPR gene editing
format results into:
- recent discoveries
- clinical applications
- upcoming trials
sites: nature.com, science.org, nih.gov
time: past year
Market Intelligence with Structured Output
/perplexity analyze the AI chip market
include:
- major players
- market share
- new developments
return as structured JSON
Current Events Analysis
/perplexity what are the key climate policy changes
format as:
- new regulations
- country commitments
- industry response
time: past month
Technical Deep Dives
/perplexity using sonar-pro: explain advances in quantum error correction
focus on:
- new methods
- experimental results
- practical implications
File-Based Research
/perplexity analyze this market report and find similar recent market trends
[attach your PDF/DOCX file]
Remember that the command can handle natural language queries, so feel free to ask questions as you would to a research assistant. The more specific and well-structured your query, the better the results will be.
