DEV Community

Alex Spinov
Alex Spinov

Posted on

Perplexity Has a Free API: Build AI Search Into Your Applications

#ai #api #python #search

What is Perplexity API?

Perplexity API gives you programmatic access to their AI search engine — the one that provides real-time, cited answers from the web. Unlike ChatGPT which hallucinates sources, Perplexity actually searches the internet and provides verifiable citations.

Why Perplexity API?

  • Real-time web search — answers include current information, not training data
  • Built-in citations — every claim comes with a source URL
  • OpenAI-compatible — drop-in replacement for OpenAI SDK
  • Multiple models — sonar-small (fast), sonar-medium, sonar-large (best)
  • No RAG needed — the web IS your knowledge base

Quick Start 

from openai import OpenAI

client = OpenAI(
    api_key="your-perplexity-key",  # From pplx.ai/settings/api
    base_url="https://api.perplexity.ai"
)

response = client.chat.completions.create(
    model="sonar",
    messages=[{
        "role": "user",
        "content": "What are the latest Kubernetes security best practices in 2026?"
    }]
)

print(response.choices[0].message.content)
# Returns answer WITH citations to real, current sources
Enter fullscreen mode Exit fullscreen mode

Search with Citations 

response = client.chat.completions.create(
    model="sonar",
    messages=[{
        "role": "user",
        "content": "Compare Terraform vs Pulumi vs Crossplane for IaC in 2026"
    }],
    return_citations=True
)

# Access citations
for citation in response.citations:
    print(f"Source: {citation.url}")
Enter fullscreen mode Exit fullscreen mode

Streaming 

stream = client.chat.completions.create(
    model="sonar",
    messages=[{"role": "user", "content": "Latest Docker security vulnerabilities 2026"}],
    stream=True
)

for chunk in stream:
    if chunk.choices[0].delta.content:
        print(chunk.choices[0].delta.content, end="")
Enter fullscreen mode Exit fullscreen mode

Build a Research Agent 

def research_topic(topic, depth=3):
    results = []
    questions = [f"What is {topic} and why does it matter?"]

    for i in range(depth):
        for q in questions:
            response = client.chat.completions.create(
                model="sonar",
                messages=[{"role": "user", "content": q}]
            )
            results.append({
                "question": q,
                "answer": response.choices[0].message.content,
                "citations": getattr(response, 'citations', [])
            })

        # Generate follow-up questions
        follow_up = client.chat.completions.create(
            model="sonar",
            messages=[{
                "role": "user",
                "content": f"Based on this research about {topic}, what are 3 important follow-up questions? Return only the questions."
            }]
        )
        questions = follow_up.choices[0].message.content.split("\n")

    return results

research = research_topic("service mesh adoption in 2026")
Enter fullscreen mode Exit fullscreen mode

Perplexity vs Alternatives

Feature Perplexity API OpenAI Google Search API Tavily
Real-time search Built-in No Raw results Yes
AI answers Yes Yes (no search) No Yes
Citations Automatic None Links only Yes
API compatibility OpenAI SDK Native REST REST
Pricing $0.20-1/1K req $30/M tok $5/1K queries $5/1K

Real-World Use Case

A market research firm spent 40 hours per report manually searching and citing sources. They built a Perplexity-powered research pipeline: auto-search 50 queries, aggregate cited answers, generate report draft. Result: 40 hours became 2 hours, and every data point had a verifiable source — something their clients loved.

Building AI-powered research tools? I specialize in search-augmented AI systems with verified sources. Contact spinov001@gmail.com or explore my data automation on Apify.

Top comments (0)