Zenrows vs Firecrawl Benchmark

Scripts and raw data from an original benchmark comparing Zenrows and Firecrawl's /scrape endpoint across 7 target URLs, conducted in May 2026.

Companion code for the Zenrows article Best Firecrawl Alternative for Anti-Bot Bypass.

Test targets

Target	Type	URL
Amazon product page	Protected	https://www.amazon.com/ref=nav_logo
Glassdoor company page	Protected	https://www.glassdoor.com/Overview/Working-at-Google-EI_IE9079.11,17.htm
LinkedIn public profile	Protected	https://www.linkedin.com/in/satyanadella/
Google SERP results page	Protected	https://www.google.com/search?q=web+scraping+api
IKEA (Cloudflare-protected e-commerce)	Protected	https://www.ikea.com/
Zillow real estate listing	Unprotected	https://www.zillow.com/homes/for_sale/
BBC news article	Unprotected	https://www.bbc.com/news/technology

Methodology

• 100 requests per target per platform
• Zenrows configured with mode=auto on all requests
• Firecrawl using its /scrape endpoint
• Recorded per request: HTTP status code, response time (ms), and page title detection

What this test does not cover

• Bulk asynchronous job workflows
• Session persistence across multi-step logged-in workflows
• Firecrawl's crawl and map endpoints (only /scrape is exercised)

Setup

Prerequisites

• Python 3.10+
• A Zenrows API key
• A Firecrawl API key

1. Clone the repository

git clone https://github.com/ZenRows/zenrows-vs-firecrawl-benchmark.git
cd zenrows-vs-firecrawl-benchmark

2. Create a virtual environment

python -m venv .venv
source .venv/bin/activate   # macOS/Linux
.venv\Scripts\activate      # Windows

3. Install dependencies

pip install -r requirements.txt

4. Configure API keys

Both scripts read their credentials from the environment:

export ZENROWS_API_KEY=your_zenrows_api_key
export FIRECRAWL_API_KEY=your_firecrawl_api_key

Running the benchmark

Each platform is run by its own script:

python zenrows.py      # writes zenrows_benchmark_results.csv
python firecrawl.py    # writes firecrawl_benchmark_results.csv

Each run issues 100 requests per target, so expect it to take a while.

test.py is a small standalone script for confirming your Firecrawl credentials work before starting a full run.

Output

Both scripts write a CSV to the repository root recording, per request: target name, URL, HTTP status code, response time, whether a page title was detected, and success status.

The data/ folder holds the aggregated spreadsheet used to produce the figures in the article.

Repository structure

.
├── zenrows.py        # Zenrows benchmark run
├── firecrawl.py      # Firecrawl /scrape benchmark run
├── test.py           # quick Firecrawl credential check
├── data/             # aggregated results used in the article
├── requirements.txt
└── README.md

License

MIT