ParadeDB is an Apache-2.0-licensed PostgreSQL extension that you can install, use, and modify for free.
Yes. ParadeDB is released under the permissive Apache 2.0 license. You pay nothing to download, compile, or install the extension in any PostgreSQL database, including production environments.
The license grants you rights to use, modify, distribute, and even sell software that includes ParadeDB—as long as you keep the license notice and provide attribution. No copyleft clauses apply.
No hard limits exist. ParadeDB’s full-text, vector, and hybrid search features are available in the open-source repository. Commercial support or hosting services may cost money, but the code itself remains free.
Running ParadeDB on large datasets can increase compute and storage bills. If you choose managed PostgreSQL hosting, your provider—not ParadeDB—charges for resources.
Clone the GitHub repo, build the extension, copy the shared library to PostgreSQL’s lib
directory, and run CREATE EXTENSION paradedb;
from psql. Details appear in the syntax section below.
After installation, you can store product vectors or tsvectors and query them with ParadeDB’s parade_match
function to power search in an ecommerce app.
Pin the exact ParadeDB version in production, monitor index sizes, and benchmark query latency. Contribute bug fixes back to the project to keep the ecosystem healthy.
See the dedicated section below for pitfalls such as forgetting the CREATE EXTENSION
step or mismatching PostgreSQL versions.
No. The Apache 2.0 license covers commercial use without extra fees.
Yes. You may fork, modify, and even distribute a closed-source build, provided you retain the original license notice.
Indexes consume RAM and disk. Benchmark on production-like data, tune work_mem, and consider partitioning large tables.