Pandas vs DuckDB vs Polars: I Ran 20 Million Rows Through All Three. Here's What Actually Happened.

The Data Bytes

SubscribeSign in

Pandas vs DuckDB vs Polars: I Ran 20 Million Rows Through All Three. Here's What Actually Happened.<br>Alright alright alright, let’s do the benchmarking exercise for every data engineer’s favourite libraries in the market.

Data bytes<br>May 23, 2026<br>∙ Paid

Share

If you work with data in Python, you have probably heard at least one of these arguments:<br>“Polars is 10x faster than Pandas.”<br>“DuckDB is just SQL — it can’t compete on Python workflows.”<br>“Pandas is slow but it’s what everyone knows, so who cares.”

Everyone has an opinion. Almost nobody has a table.<br>Data’s Substack is a reader-supported publication. To receiv…

Subscribe

Continue reading this post for free, courtesy of Data bytes.<br>Claim my free post<br>Or purchase a paid subscription.

© 2026 Data bytes · Privacy ∙ Terms ∙ Collection notice<br>Start your SubstackGet the app<br>Substack is the home for great culture

This site requires JavaScript to run correctly. Please turn on JavaScript or unblock scripts