{"data":{"slug":"polars","name":"Polars","tagline":"DataFrames that respect your CPU","category":"python","categoryName":"Python","url":"https://pola.rs","github":"https://github.com/pola-rs/polars","pricing":"open-source","hook":"Polars is what pandas would be if designed today.","whatsGood":"The API is expressive: method chaining, lazy queries, and automatic parallelization. Memory efficiency is dramatically better than pandas for large datasets. The query optimizer makes naive code fast.","theCatch":"Different API than pandas—existing codebases need rewriting. The ecosystem is smaller; some pandas integrations do not exist.","verdict":"What pandas would be if designed today.","addedAt":"2026-08-18","featured":false,"domain":"Python","subSpecialty":"Data","capabilities":["dataframes"],"surfaces":["Self-host"],"ecosystem":["Python"],"licenseModel":"OSS/MIT","githubStars":39619},"related":[{"slug":"fastapi","name":"FastAPI","tagline":"Modern Python APIs at startup speed"},{"slug":"pydantic","name":"Pydantic","tagline":"Data validation using Python type hints"},{"slug":"uv","name":"uv","tagline":"Python packaging at Rust speed"},{"slug":"ruff","name":"Ruff","tagline":"Python linter faster than your test suite"},{"slug":"typer","name":"Typer","tagline":"CLI apps with type hints and zero effort"}],"links":{"self":"/api/gems/polars","html":"/gems/polars","markdown":"/gems/polars.md","category":"/api/gems?category=python","categoryPage":"/stacks/python"}}