GeoDesk

GeoDesk is a free, open-source spatial database engine built around OpenStreetMap data. It is best for developers and analysts who need compact, self-hosted OSM storage and fast feature queries in Java, Python or C++. Choose it for OSM-specific performance and tooling; look elsewhere if you need 3D geometry or a general-purpose database platform.

Overview

GeoDesk stores OpenStreetMap features in single-file Geographic Object Libraries (GOLs), rather than in a conventional server database. The design keeps OSM concepts such as relations intact and targets applications, map products and analysis pipelines that work directly with OSM data. It is self-hosted, with no hosted service or subscription to manage.

The project comes from an independent team focused on cartography and geospatial work. GeoDesk states that it is not affiliated with, endorsed by or sponsored by OpenStreetMap, so users should treat it as an independent toolchain for OSM data.

Key features

The GOL command-line utility imports OpenStreetMap PBF files into the single-file format. GeoDesk says a GOL is typically about 40 percent larger than its PBF source, while remaining a small fraction of the footprint of a traditional database. That makes a library straightforward to move or archive, although it does not minimize storage as aggressively as the source PBF itself.

GeoDesk reports that its most common spatial queries run 50 times faster than equivalent SQL queries. The benefit is aimed at OSM-style feature and tag lookups, not every database workload. GOQL is a concise query language based on feature types and tags and resembles MapCSS, giving OSM users a focused alternative to SQL. A GOL can also become a compressed tile repository, so a downloader can retrieve only the regions required instead of a complete dataset.

For geometry processing, documented integrations connect Java Topology Suite and Python Shapely to operations such as buffers, unions and triangulation. GeoDesk’s Python Map feature renders queried features in a Leaflet-based map that can open in a browser or be saved as HTML. Coordinate transforms are supported. The spatial type set covers points, lines, rings, polygons, multi-geometries and geometry collections, but not 3D geometry.

Developer tooling and guides cover Java, Python and C++, alongside the GOL utility. Python requires version 3.7 or newer. GeoDesk says it performs well on systems capable of running 64-bit Java or Python, while the C++ toolkit documentation covers 64-bit Windows, Linux and macOS. Building that C++ library requires a C++20-capable compiler and CMake 3.14 or later, a meaningful setup constraint for older build environments. Documentation, tutorials, GitHub repositories and API references provide the main support path.

Pricing

GeoDesk costs nothing. The GeoDesk plan is 100% free and open-source, with self-hosted deployment. There is no free trial because the product is already offered as a free plan, and no seat, usage or renewal charge is stated. This suits individuals, research projects and teams that can run and maintain their own tools; it does not provide a paid hosted tier for readers who want managed operations.

Platforms

GeoDesk supports Linux, macOS and Windows. The documented C++ toolkit specifically lists 64-bit versions of all three. Java and Python tooling broadens the practical choice of development environment, subject to the Python 3.7-or-newer requirement and the need for 64-bit Java or Python-capable hardware.

Who it's for

Choose GeoDesk for software that builds on OpenStreetMap’s feature model, repeated spatial lookups, regional data distribution or geospatial analysis. It is particularly compelling when a portable file and OSM-aware relations matter more than SQL compatibility. Analysts can combine GOQL with Shapely or Java Topology Suite, and developers can publish browser-viewable Leaflet maps from Python.

It is a poor fit for systems that require 3D geometry, a general relational database, or a managed cloud service. Teams standardizing on SQL will also need to learn GOQL and adapt workflows around GOL files.

Pros and cons

• Pro: Single-file GOLs preserve OSM relations and remain far smaller than a traditional database, simplifying distribution and local deployment.
• Pro: GeoDesk reports up to 50-times-faster common spatial queries than SQL equivalents for workloads matching its feature-and-tag model.
• Pro: Java, Python and C++ libraries, a command-line importer and documented geometry integrations support several development stacks.
• Pro: Compressed tile repositories allow regional downloads instead of requiring an entire dataset.
• Con: GOLs are usually about 40 percent larger than the originating PBF files, so they are not the smallest possible OSM representation.
• Con: The engine has no 3D geometry support and its OSM-focused query model is not a drop-in replacement for SQL.
• Con: C++ users need a C++20 compiler and CMake 3.14 or newer, which can complicate older build systems.

Alternatives

Spatial Databases is the broader category to survey when you want to compare engines beyond GeoDesk’s OSM specialization.

Azure SQL Database is a better choice when you need a cloud SQL service, broad platform access and a free offer with 100,000 vCore seconds, 32 GB of data storage and 32 GB of backup storage across up to 10 databases per subscription.

CrateDB fits teams wanting a distributed database with a free shared tier of 2 vCPUs, 2 GiB RAM and 8 GiB storage rather than a local OSM file engine.

MySQL is the conventional choice for SQL and NoSQL applications, replication and high-availability components through its Community Edition.

RavenDB suits users seeking a document database with developer and community deployment options.

Supabase is preferable when a hosted development platform and services such as database, authentication-oriented usage and file storage are more important than OSM-native storage.

DuckDB UI offers a browser-based local SQL notebook with optional MotherDuck connectivity for readers who want an interactive SQL interface.

Microsoft SQL Server is the established SQL Server route for teams that need its enterprise or standard editions.

OpenSearch is a free Apache 2.0 option when open-source search and analytics are the priority.

Verdict

GeoDesk is a strong recommendation for developers and analysts building OSM-centric applications that benefit from portable files, OSM relations and fast spatial lookups. Its free, open-source self-hosted model removes licensing cost and its multi-language tooling is practical. Pick another database when you need 3D data, SQL-first interoperability or managed hosting.