CrUX Cache

Cached Chrome User Experience Report (CrUX) data for global and country-specific origins with popularity rankings.

Download Data

Option 1: Command Line (Quick & Easy)

Download datasets directly from your terminal with a single command:

# Download latest global dataset
bash <(curl -sSL https://raw.githubusercontent.com/lonetis/crux-cache/main/download.sh) global

# Download latest country-specific dataset (e.g., us, de, jp)
bash <(curl -sSL https://raw.githubusercontent.com/lonetis/crux-cache/main/download.sh) us

# Download specific month (YYYYMM format)
bash <(curl -sSL https://raw.githubusercontent.com/lonetis/crux-cache/main/download.sh) global 202510
bash <(curl -sSL https://raw.githubusercontent.com/lonetis/crux-cache/main/download.sh) us 202509

This will download and merge all chunks into a single CSV file (e.g., crux_global_202510.csv or crux_us_202509.csv) in your current directory.

Available datasets: Check lonetis.github.io/crux-cache or view data/datasets.json for the current list of available datasets.

Option 2: Website

Visit lonetis.github.io/crux-cache to browse and download any dataset with a streaming browser interface

Option 3: Python Package

Install and use programmatically:

pip install crux-cache

from crux_cache import CruxCache

cache = CruxCache()

# Iterate over all domains in the latest global dataset
for origin, rank in cache.get_dataset('global'):
    print(f"{origin}: {rank}")

# Get top 1k domains from a specific month
for origin, rank in cache.get_dataset('us', month='202510', max_rank=1000):
    print(f"{origin}: {rank}")

See python/README.md for complete documentation, API reference, and examples.

Data Format

CSV with two columns:

• origin: Full URL (e.g., https://www.google.com)
• rank: Popularity bucket (1000, 10000, 100000, 1000000, etc.)

Example:

origin,rank
https://www.google.com,1000
https://www.youtube.com,1000
https://www.facebook.com,1000

Dataset Details

• Scope: All ~18M origins tracked by CrUX
• Updates: Monthly (automated)
• Format: CSV chunks (25MB each, uncompressed)
• Source: Chrome User Experience Report via BigQuery

How It Works

Data is stored as chunked CSV files (25MB each) for GitHub compatibility. Only the first chunk contains a CSV header, allowing seamless concatenation.

Both download methods (CLI and website) automatically merge chunks into a single CSV file for you.

License

MIT License - See LICENSE

CrUX data provided by Google under CrUX Dataset Terms