---
title: 0.18.0
---

## Performance Improvements 🚀

- Avoid re-querying Top N iterators that have no more results (up to 1.5X speedup)
- Lazily load bitpacked columns (up to 2X speedup)
- Use [Arrow](https://arrow.apache.org/) for fast field execution (up to 1.6X speedup)
- Skip computing scores if not necessary during Top N queries

## Aggregate Query Pushdown

We're making further progress in our roadmap to [push down aggregate queries](/welcome/roadmap#analytics). This release
pushes down bucketing queries with `GROUP BY` or `GROUP BY...ORDER BY`.

## New SQL UX

We are making progress on [v2 of the new SQL UX](/v2/overview). While not completed, `0.18.0` brings fuzzy, boost, and query builder function
support to v2. The new UX is still in beta and we would appreciate any community testing/feedback.

## Stability Improvements 💪

- Various correctness fixes to Top N queries
- Custom scans over partitioned tables don't error if one of the partitioned tables is empty

The full changelog is available [here](https://github.com/paradedb/paradedb/releases/tag/v0.18.0).