Perplexity Deploy Integration

Overview

Deploy applications using Perplexity's AI search API (api.perplexity.ai). Perplexity uses an OpenAI-compatible chat completions format with real-time web search grounding.

Prerequisites

• Perplexity API key stored in PERPLEXITY_API_KEY environment variable
• Application using fetch or OpenAI-compatible SDK
• Platform CLI installed (vercel, docker, or gcloud)

Instructions

Step 1: Configure Secrets

# Vercel
vercel env add PERPLEXITY_API_KEY production

# Cloud Run
echo -n "your-key" | gcloud secrets create perplexity-api-key --data-file=-

Step 2: Deploy Search API with Streaming

// api/search.ts
export const config = { runtime: "edge" };

export default async function handler(req: Request) {
  const { query, model } = await req.json();

  const response = await fetch("https://api.perplexity.ai/chat/completions", {
    method: "POST",
    headers: {
      "Authorization": `Bearer ${process.env.PERPLEXITY_API_KEY}`,
      "Content-Type": "application/json",
    },
    body: JSON.stringify({
      model: model || "sonar",
      messages: [{ role: "user", content: query }],
      stream: true,
      return_citations: true,
    }),
  });

  return new Response(response.body, {
    headers: { "Content-Type": "text/event-stream" },
  });
}

Step 3: Docker with Cache Layer

import { Redis } from "ioredis";

const redis = new Redis(process.env.REDIS_URL!);

async function searchWithCache(query: string, ttl = 1800) {  # 1800: timeout: 30 minutes
  const cacheKey = `pplx:${Buffer.from(query).toString("base64")}`;
  const cached = await redis.get(cacheKey);
  if (cached) return JSON.parse(cached);

  const response = await fetch("https://api.perplexity.ai/chat/completions", {
    method: "POST",
    headers: {
      "Authorization": `Bearer ${process.env.PERPLEXITY_API_KEY}`,
      "Content-Type": "application/json",
    },
    body: JSON.stringify({
      model: "sonar",
      messages: [{ role: "user", content: query }],
      return_citations: true,
    }),
  });

  const result = await response.json();
  await redis.set(cacheKey, JSON.stringify(result), "EX", ttl);
  return result;
}

Step 4: Vercel Configuration

{
  "functions": {
    "api/search.ts": { "maxDuration": 30 }
  }
}

Step 5: Health Check

export async function GET() {
  try {
    const response = await fetch("https://api.perplexity.ai/chat/completions", {
      method: "POST",
      headers: {
        "Authorization": `Bearer ${process.env.PERPLEXITY_API_KEY}`,
        "Content-Type": "application/json",
      },
      body: JSON.stringify({
        model: "sonar",
        messages: [{ role: "user", content: "ping" }],
        max_tokens: 1,
      }),
    });
    return Response.json({ status: response.ok ? "healthy" : "degraded" });
  } catch {
    return Response.json({ status: "unhealthy" }, { status: 503 });  # HTTP 503 Service Unavailable
  }
}

Error Handling

Issue	Cause	Solution
Rate limited	Too many requests	Cache responses, use queue
Stale search results	Cached too long	Reduce cache TTL for time-sensitive queries
API key invalid	Key expired	Regenerate at perplexity.ai settings
Stream interrupted	Network timeout	Implement reconnection logic

Examples

Basic usage: Apply perplexity deploy integration to a standard project setup with default configuration options.

Advanced scenario: Customize perplexity deploy integration for production environments with multiple constraints and team-specific requirements.

Resources

• Perplexity API Documentation
• Perplexity Models

Next Steps

For multi-environment setup, see perplexity-multi-env-setup.

Output

• Configuration files or code changes applied to the project
• Validation report confirming correct implementation
• Summary of changes made and their rationale