Open geographic datasets including crowdsourced maps, place name databases, satellite imagery, and administrative boundary data.
5sources
Geospatial and mapping data describes where things are: roads, buildings, borders, place names, coastlines, and points of interest. It matters for a RAG (Retrieval-Augmented Generation) system whenever a question has a location in it, from “what is near this address” to “which region does this council cover”. Geographic data also gives you a reliable backbone of place names and administrative areas to ground answers against, which is useful even in systems that are not primarily about maps.
The main choice here is between crowdsourced and authoritative sources. Community mapping projects offer astonishing global coverage and detail, updated constantly by volunteers, but completeness varies from place to place. Curated reference datasets and official mapping agencies trade some of that richness for consistency and vetted accuracy. If you need dependable boundaries or a clean list of countries and cities, reach for the curated sets; if you need street level detail, the crowdsourced maps are hard to beat.
Two practical points. First, format: geographic data comes as shapefiles, GeoJSON, or compressed map extracts, which store shapes and coordinates rather than sentences, so you will usually convert the attributes and place descriptions into plain text before indexing them for retrieval. Second, licensing: several major sources use a share-alike licence (which requires you to release derived data under the same terms), so check the conditions before you build the data into a commercial product.
The sources below span global crowdsourced maps, curated reference layers, and open place name databases.
A geographical database of more than 11 million place names, each with coordinates, population, elevation, and administrative divisions, covering every country. It works well as a gazetteer for resolving place names to locations.
A conflated global building-footprint layer that merges Google Open Buildings V3, Microsoft Global ML Footprints, and OpenStreetMap into roughly 2.7 billion footprints, each labelled by its source. Built by VIDA and hosted on Source Cooperative in cloud-native formats, it offers more complete coverage than any single provider on its own.
A public domain map dataset built for cartography, offered at three scales: 1:10m, 1:50m, and 1:110m. It bundles cultural data (borders, cities, roads), physical data (coastlines, rivers, lakes), and raster imagery, all ready to drop into a map.
A crowdsourced map of the world, built and maintained by a community of volunteers. It holds detailed geographic data on roads, buildings, land use, points of interest, and much more, which you can download as a full planet file or as smaller regional extracts.
An open map data project run under the Linux Foundation by a group of industry members. It combines and harmonises data from OpenStreetMap, Microsoft, Meta, and others into consistent, interoperable map layers you can build on freely.