Polars 2.0 Launches With Default Streaming And Enhanced SQL Support
Polars 2.0 introduces a default streaming execution engine, first-class SQL support, and improved out-of-core processing to enhance performance for large-scale data analysis.
Polars 2.0 is out, bringing architectural changes that fix legacy design decisions and remove technical debt from previous versions.
The streaming engine is now the default execution mode. SQL support is promoted to a first-class feature. You also get initial out-of-core support that spills data to disk at memory limits, plus a new Map data type for data manipulation.
While the team initially called this a minor release, the final scope grew wider. The update includes various performance optimizations. These changes lead to claims that Polars ranks among the fastest single-node analytical SQL engines.
If you are upgrading, check the migration guide in the official documentation. The developers also provided benchmark data so you can check performance metrics and assess how the changes impact your existing workflows.