PMTiles

PMTiles is a single-file archive format for pyramids of tiled data. A PMTiles archive can be hosted on a storage platform like S3, and enables low-cost, zero-maintenance map applications.

• PMTiles is a general format for tiled data addressed by Z/X/Y coordinates. This can be cartographic basemap vector tiles, remote sensing observations, JPEG images, or more.

• PMTiles readers use HTTP Range Requests to fetch only the relevant tile or metadata inside a PMTiles archive on-demand.

• The arrangement of tiles and directories is designed to minimize the amount of overhead requests when panning and zooming.

The current specification version of PMTiles is version 3, which you can read on GitHub.

The Viewer is a browser-based tool for visualizing PMTiles on a map, inspecting metadata, and debugging individual tiles.

You can access the Viewer at protomaps.github.io/PMTiles/.

The official tool for reading and writing PMTiles is the pmtiles command line utility.

Download pre-compiled binaries for your platform at GitHub Releases.

Examples and source code for pmtiles are in the README for go-pmtiles on GitHub.

PMTiles is designed for being read directly in web browsers in conjunction with a JavaScript map library.

See the docs on viewing PMTiles in Leaflet, MapLibre GL JS and OpenLayers.

pmtiles is a python package for reading and writing PMTiles. GitHub and via the Python Package Index.

Documentation is available on GitHub.

Protomaps Downloads is a free service to download up-to-date basemap tilesets from OpenStreetMap.

We'll focus on a small area of Berkeley, CA, but you can follow these instructions for any area. Use the search box and the rectangle or polygon tool to draw the area you're interested in:

You'll need to use any local HTTP server that supports Range Requests to test the file locally. the npm package http-server is a popular one:

npm install -g http-server
# from the directory containing your .pmtiles file
http-server . --cors
# your file is now available at http://localhost:8080/

You can also use the HTTP server built into your frontend framework, usually by placing the .pmtiles file into the /public folder.

See the Leaflet or MapLibre documentation for how to load the PMTiles into a map rendering frontend library.

the pmtiles command line tool converts MBTiles to PMTiles with this command:

pmtiles convert INPUT.mbtiles OUTPUT.pmtiles

Using a file from OSGeo's GeoTIFF samples and the rio-mbtiles Python tool:

# convert single-band to 3-band RGB GeoTIFF
gdal_translate -expand rgb input.tif input_rgb.tif

# in this example create 512x512 PNG tiles from zooms 0 to 16.
rio mbtiles input_rgb.tif output.mbtiles --format PNG --zoom-levels 0..16 --tile-size 512 --resampling bilinear

pmtiles convert output.mbtiles output.pmtiles

Tippecanoe is the supported tool for converting datasets into tiles. Tippecanoe version 2.17 and later supports direct PMTiles output.

An example of how to convert a Shapefile: US Census Bureau Tabulation Areas in two steps, using GDAL's ogr2ogr command line tool:

ogr2ogr -t_srs EPSG:4326 cb_2018_us_zcta510_500k.json cb_2018_us_zcta510_500k.shp
# Creates a layer in the vector tiles named "zcta"
tippecanoe -zg --projection=EPSG:4326 -o cb_2018_us_zcta510_500k_nolimit.pmtiles -l zcta cb_2018_us_zcta510_500k.json

• Deploy your tileset to the Internet via Cloud Storage.

• Edit OpenStreetMap and download basemaps to get new data. Your improvements to the map are available for anyone!

• Learn how to customize your map's Basemap Layers.