# GeoTessera R library now available

Date: 2026-03-16
Author: Anil Madhavapeddy
Canonical URL: https://geotessera.org/blog/2026-03-16-geotessera-r-library

# GeoTessera R library available

[Simon Frost](https://www.microsoft.com/en-us/research/people/sdwfrost/) from Microsoft Research has built an [R port](https://lassa-sentinel.github.io/GeoTessera/) of the [geotessera](https://pypi.org/project/geotessera/) library. The new [GeoTessera R package](https://lassa-sentinel.github.io/GeoTessera/) makes it easy for R users to integrate 10m-resolution geospatial embeddings into their existing workflows.

The R library mirrors the functionality of the Python package, letting you retrieve numpy tiles, extract embeddings, and export data from R:

```r
library(GeoTessera)
gt <- geotessera()
tiles <- gt$get_tiles(bbox = c(-0.2, 51.4, 0.1, 51.6), year = 2024)
gt$export_embedding_geotiffs(tiles = tiles, output_dir = "london_tiles")
```

The package builds on R geospatial libraries including `sf`, `terra`, and `arrow`, and follows the R6 class conventions familiar to R users.

We welcome other community contributions in any language. The underlying data is currently served over HTTP as numpy files, but we are curently porting to [Zarr v3 over HTTP](https://anil.recoil.org/notes/tessera-zarr-v3-layout). Supporting either is fine depending on your needs.

## Get started

Install the R package from GitHub:

```r
remotes::install_github("lassa-sentinel/GeoTessera")
```

Full documentation is available at [lassa-sentinel.github.io/GeoTessera](https://lassa-sentinel.github.io/GeoTessera/).

If you're building tools or libraries on top of TESSERA embeddings, we'd love to hear about it — come say hello on [Zulip](https://eeg.zulipchat.com) or open an issue or discussion on [GitHub](https://github.com/ucam-eo/geotessera).