Tech Bytes Logo

Tech Bytes Developer Hub

September 28, 2025

🚀 API LAUNCH

Perplexity Search API: Real-Time AI Search for Developers

Build intelligent search experiences with Perplexity's new API - combining real-time web results with AI-powered understanding

🔍 Real-time Search ⚡ Low Latency 💻 Developer-Friendly
Dillip Chowdary

Dillip Chowdary

API & Developer Tools Expert

Perplexity has officially launched its Search API, empowering developers to integrate advanced AI-powered search capabilities directly into their applications. This API combines real-time web search with Perplexity's renowned answer engine, delivering accurate, contextual results with source citations.

Key Features of Perplexity Search API

Core Capabilities

  • Real-time Web Search: Up-to-date information from across the web
  • AI-Powered Answers: Natural language responses with context
  • Source Citations: Every answer includes verifiable sources
  • Multi-Modal Support: Text, code, and structured data responses

Technical Specifications

  • 🚀 Latency: <500ms average response time
  • 🔧 Rate Limits: 1000 requests/minute (Pro tier)
  • 🌍 Languages: 20+ languages supported
  • 💰 Pricing: $5 per 1000 requests

API Endpoints

Base URL

https://api.perplexity.ai/v1

Available Endpoints

POST /search

Perform a search query with AI-powered results

POST /chat/completions

Interactive chat with search capabilities

GET /models

List available search models

Quick Start Code Examples

Python Example

import requests
import json

# Your API key from Perplexity dashboard
API_KEY = "pplx-YOUR_API_KEY_HERE"

# API endpoint
url = "https://api.perplexity.ai/v1/search"

# Headers with authentication
headers = {
    "Authorization": f"Bearer {API_KEY}",
    "Content-Type": "application/json"
}

# Search query
data = {
    "model": "pplx-7b-online",
    "messages": [
        {
            "role": "user",
            "content": "What are the latest developments in quantum computing?"
        }
    ],
    "max_tokens": 1000,
    "temperature": 0.7,
    "return_citations": True,
    "search_domain_filter": ["arxiv.org", "nature.com", "science.org"]
}

# Make the request
response = requests.post(url, headers=headers, json=data)

# Parse and print the response
if response.status_code == 200:
    result = response.json()
    print("Answer:", result["choices"][0]["message"]["content"])
    print("\nCitations:")
    for citation in result["citations"]:
        print(f"- {citation['title']}: {citation['url']}")
else:
    print(f"Error: {response.status_code}")
    print(response.text)

JavaScript/Node.js Example

const axios = require('axios');

// Your API key
const API_KEY = 'pplx-YOUR_API_KEY_HERE';

async function searchPerplexity(query) {
    const url = 'https://api.perplexity.ai/v1/search';

    const config = {
        headers: {
            'Authorization': `Bearer ${API_KEY}`,
            'Content-Type': 'application/json'
        }
    };

    const data = {
        model: 'pplx-7b-online',
        messages: [
            {
                role: 'user',
                content: query
            }
        ],
        max_tokens: 1000,
        temperature: 0.7,
        return_citations: true,
        search_recency_filter: '24h'  // Results from last 24 hours
    };

    try {
        const response = await axios.post(url, data, config);

        console.log('Answer:', response.data.choices[0].message.content);
        console.log('\nCitations:');
        response.data.citations.forEach(citation => {
            console.log(`- ${citation.title}: ${citation.url}`);
        });

        return response.data;
    } catch (error) {
        console.error('Error:', error.response?.data || error.message);
        throw error;
    }
}

// Example usage
searchPerplexity('Latest AI breakthroughs in 2025')
    .then(result => console.log('Search completed'))
    .catch(err => console.error('Search failed:', err));

cURL Example

curl -X POST https://api.perplexity.ai/v1/search \
  -H "Authorization: Bearer pplx-YOUR_API_KEY_HERE" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "pplx-7b-online",
    "messages": [
      {
        "role": "user",
        "content": "Explain the latest CRISPR gene editing techniques"
      }
    ],
    "max_tokens": 1000,
    "temperature": 0.7,
    "return_citations": true,
    "search_domain_filter": ["pubmed.gov", "nature.com"]
  }'

Advanced Features

Domain Filtering

Restrict search results to specific domains for more targeted results:

"search_domain_filter": ["github.com", "stackoverflow.com"]

Recency Filtering

Get only recent results with time-based filtering:

"search_recency_filter": "24h" // Options: 24h, week, month, year

Streaming Responses

Enable streaming for real-time response generation:

"stream": true // Returns Server-Sent Events

Context Window

Maintain conversation context across multiple queries:

"context_window": 4000 // Number of tokens to maintain

Real-World Use Cases

🔬 Research Applications

  • • Academic paper discovery and summarization
  • • Patent search and analysis
  • • Literature reviews with citations
  • • Trend analysis in scientific fields

💼 Business Intelligence

  • • Competitor analysis and monitoring
  • • Market research and insights
  • • Customer sentiment analysis
  • • Industry news aggregation

🤖 AI Applications

  • • Chatbots with real-time information
  • • Virtual assistants with web access
  • • Automated content generation
  • • Fact-checking systems

📱 Developer Tools

  • • Code search and documentation
  • • Error resolution with Stack Overflow
  • • API documentation discovery
  • • Package vulnerability checking

Pricing Plans

Free Tier

$0/month

  • ✓ 100 requests/month
  • ✓ Basic search model
  • ✓ 5 requests/minute
  • ✓ Community support
Get Started

Pro

$200/month

  • ✓ 40,000 requests/month
  • ✓ All search models
  • ✓ 1000 requests/minute
  • ✓ Priority support
  • ✓ Advanced filtering
Start Pro Trial

Enterprise

Custom

  • ✓ Unlimited requests
  • ✓ Custom models
  • ✓ No rate limits
  • ✓ Dedicated support
  • ✓ SLA guarantee
  • ✓ On-premise option
Contact Sales

Getting Started in 5 Minutes

  1. 1
    Sign up at Perplexity

    Visit perplexity.ai/api and create your account

  2. 2
    Get your API key

    Navigate to Settings → API Keys and generate a new key

  3. 3
    Install SDK (optional)

    pip install perplexity-python or npm install perplexity-js

  4. 4
    Make your first request

    Copy one of the code examples above and replace the API key

  5. 5
    Build something amazing!

    Start integrating AI-powered search into your applications

Ready to Build with AI Search?

Join thousands of developers using Perplexity's Search API to power intelligent applications.

Related Articles