Firecrawl

Search

Web scraping, crawling, and structured data extraction for LLMs.

78.9
out of 100

Moderate accuracy (68.89%) — some responses were incomplete or partially correct. Very reliable — 9/9 requests got a response. Median response time: 13499ms (p95: 14284ms).

Last tested: 3/22/2026, 5:12:59 PM

What AI models say

GPT-4o-mini

Web scraping and data extraction with Firecrawl can yield good results, especially for straightforward tasks like scraping a standard news website, where I got perfect quality. However, the overall success rate was disappointing at only 67%, with slow response times often exceeding 10 seconds, and frequent errors like HTTP 400 on various requests suggest the service needs to improve reliability and speed for a better developer experience.

Haiku

Firecrawl extracts structured data from websites reliably enough for production use when it works, but a 67% success rate and vague HTTP 400 errors on certain tasks make it feel half-baked—you'll spend time debugging edge cases instead of building. The 10+ second latencies are acceptable for batch jobs, but the inconsistency (some tests completely undefined) suggests the API still has stability issues they need to iron out.

Accuracy (40%)
68.9
Reliability (30%)
100.0
Latency (20%)
56.9
Cost Efficiency (10%)
100.0

On-Chain Activityvia Tempo Network

This service shares a payment wallet with 12 other services on Tempo Network. Stats below reflect aggregate network activity.

24h Volume
$0.00
7d: $0.00 / All: $79.78
Unique Agents
172
Payment Channels
10484
Settlements (24h)
0
7d: 0 / All: 1000
Active since 3/18/2026 · Last active 5/10/2026
View on Tempo Explorer

Test Examples

Real requests we sent and the responses we received.

Scrape a standard news website homepage

POST /v1/scrapetypical13482ms
100%

{"data":{"markdown":"| | | |\n| --- | --- | --- |\n| | | | |\n| --- | --- | --- |\n| [![](https://news.ycombinator.com/y18.svg)](https://news.ycombinator.com/) | **[Hacker News

Crawl a documentation site with depth limit

POST /v1/crawltypical13499ms
100%

{"id":"019d1688-3e09-76bd-9a98-6c2aeadf5f6e","url":"https://api.firecrawl.dev/v1/crawl/019d1688-3e09-76bd-9a98-6c2aeadf5f6e","success":true}

Scrape a single-page application with special Unicode characters in content

POST /v1/scrapeedge748ms
80%

{"data":{"markdown":"# Example Domain\n\nThis domain is for use in documentation examples without needing permission. Avoid use in operations.\n\n[Learn more](https://iana.org/domains/example)","metad

undefined

POST /undefined13841ms
30%

HTTP 400

Service Details

API AccessMPP Directory3rd Party
POST /v1/scrape$0.002000Scrape a URL
POST /v1/crawl$0.005000Crawl a website
POST /v1/map$0.002000Map website URLs
POST /v1/search$0.004000Search the web
POST /v1/extract$0.005000Extract structured data
Endpoint: https://firecrawl.mpp.tempo.xyz
Status: active
Discovered: 3/18/2026
Price per request: $0.0020 USDC
Tests run: 9
Last tested: 3/22/2026

Test History

3/22/2026, 5:11:18 PM
score: 69%p50: 13499ms

Embed this badge

MPPrimo rating for Firecrawl<a href="https://mpprimo.com/service/044286a7-16bc-422b-b8d4-383dd0524cf7"><img src="https://mpprimo.com/api/badge/044286a7-16bc-422b-b8d4-383dd0524cf7" alt="MPPrimo rating"></a>