Firecrawl Rate Limits

Firecrawl Rate Limits

Overview

Firecrawl enforces rate limits per API key measured in requests per minute and concurrent connections. When exceeded, the API returns 429 Too Many Requests with a Retry-After header. This skill covers backoff strategies, request queuing, and proactive throttling.

Prerequisites

• Provider limits confirmed for the current account, plus an approved concurrency, budget, and exception-queue owner.
• Aggregate queue and throttle telemetry; never place scraped content or keys in retry logs.
• Synthetic jobs for testing pause, replay, and duplicate suppression behavior.

Output

Produce a rate-control receipt with policy version, concurrency, retry bound, throttle count, queue age, idempotency result, escalation owner, and any manual disposition. Stop or reduce load before a budget or safety threshold is exceeded.

Rate Limit Tiers

Concurrent crawl jobs count against concurrency limits. If the queue is full, new jobs are rejected with 429.

Instructions

Step 1: Exponential Backoff with Jitter

import FirecrawlApp from "@mendable/firecrawl-js";

const firecrawl = new FirecrawlApp({
  apiKey: process.env.FIRECRAWL_API_KEY!,
});

async function withBackoff<T>(
  operation: () => Promise<T>,
  config = { maxRetries: 5, baseDelayMs: 1000, maxDelayMs: 32000 }
): Promise<T> {
  for (let attempt = 0; attempt <= config.maxRetries; attempt++) {
    try {
      return await operation();
    } catch (error: any) {
      if (attempt === config.maxRetries) throw error;

      const status = error.statusCode || error.status;
      // Only retry on 429 (rate limit) and 5xx (server error)
      if (status && status !== 429 && status < 500) throw error;

      // Exponential delay with random jitter to prevent thundering herd
      const exponentialDelay = config.baseDelayMs * Math.pow(2, attempt);
      const jitter = Math.random() * 500;
      const delay = Math.min(exponentialDelay + jitter, config.maxDelayMs);

      console.warn(`Rate limited (${status}). Retry ${attempt + 1}/${config.maxRetries} in ${delay.toFixed(0)}ms`);
      await new Promise(r => setTimeout(r, delay));
    }
  }
  throw new Error("Unreachable");
}

// Usage
const result = await withBackoff(() =>
  firecrawl.scrapeUrl("https://example.com", { formats: ["markdown"] })
);

Step 2: Queue-Based Rate Limiting with p-queue

import PQueue from "p-queue";

// Limit to 5 concurrent requests, max 10 per second
const scrapeQueue = new PQueue({
  concurrency: 5,
  interval: 1000,
  intervalCap: 10,
});

async function queuedScrape(url: string) {
  return scrapeQueue.add(() =>
    withBackoff(() =>
      firecrawl.scrapeUrl(url, { formats: ["markdown"] })
    )
  );
}

// Scrape many URLs respecting rate limits
const urls = ["https://a.com", "https://b.com", "https://c.com"];
const results = await Promise.all(urls.map(url => queuedScrape(url)));
console.log(`Queue: ${scrapeQueue.pending} pending, ${scrapeQueue.size} queued`);

Step 3: Proactive Throttling (Pre-emptive)

class RateLimitTracker {
  private requestTimes: number[] = [];
  private windowMs: number;
  private maxRequests: number;

  constructor(maxRequests = 50, windowMs = 60000) {
    this.maxRequests = maxRequests;
    this.windowMs = windowMs;
  }

  async waitIfNeeded(): Promise<void> {
    const now = Date.now();
    this.requestTimes = this.requestTimes.filter(t => now - t < this.windowMs);

    if (this.requestTimes.length >= this.maxRequests) {
      const oldestInWindow = this.requestTimes[0];
      const waitMs = this.windowMs - (now - oldestInWindow) + 100;
      console.log(`Proactive throttle: waiting ${waitMs}ms to stay under ${this.maxRequests} RPM`);
      await new Promise(r => setTimeout(r, waitMs));
    }

    this.requestTimes.push(Date.now());
  }
}

const throttle = new RateLimitTracker(50, 60000); // 50 requests per minute

async function throttledScrape(url: string) {
  await throttle.waitIfNeeded();
  return firecrawl.scrapeUrl(url, { formats: ["markdown"] });
}

Step 4: Batch Scrape for Efficiency

// batchScrapeUrls is more efficient than individual scrapes
// It handles internal rate limiting and is cheaper on credits
const urls = [
  "https://example.com/page1",
  "https://example.com/page2",
  "https://example.com/page3",
];

// Single API call instead of 3 separate scrapes
const batchResult = await firecrawl.batchScrapeUrls(urls, {
  formats: ["markdown"],
});

console.log(`Batch scraped ${batchResult.data?.length} pages`);

Error Handling

Examples

Monitor Rate Limit Usage

class RateLimitMonitor {
  private remaining = Infinity;
  private resetAt = new Date();

  update(status: number, headers: Record<string, string>) {
    if (headers["x-ratelimit-remaining"]) {
      this.remaining = parseInt(headers["x-ratelimit-remaining"]);
    }
    if (headers["x-ratelimit-reset"]) {
      this.resetAt = new Date(parseInt(headers["x-ratelimit-reset"]) * 1000);
    }
    if (this.remaining < 5) {
      console.warn(`Low rate limit: ${this.remaining} remaining, resets at ${this.resetAt.toISOString()}`);
    }
  }
}

Resources

• Firecrawl Rate Limits
• p-queue

Next Steps

For security configuration, see firecrawl-security-basics.