There is a possible using a ["Field calculator"](https://docs.qgis.org/3.16/en/docs/user_manual/processing_algs/qgis/vectortable.html#field-calculator) inside the Graphical Modeler. Let's assume there is a point layer `'test'` with some features in it, see the image below [![input][1]][1] Thereafter I created a simple model for [reverse geocoding](https://en.wikipedia.org/wiki/Reverse_geocoding) as following: [![model][2]][2] *where* a "Field calculator" possess the following parameters [![field_calc][3]][3] In the 'Formula' tab apply this expression: reverse_geocoding(y(transform($geometry,@project_crs,'EPSG:4326')), x(transform($geometry,@project_crs,'EPSG:4326'))) A custom `reverse_geocoding()` function defined as <!-- language: python --> import geopy from geopy.geocoders import Nominatim from geopy.point import Point from qgis.core import * from qgis.gui import * locator = Nominatim(user_agent="MyReverseGeocoder", timeout=10) @qgsfunction(args='auto', group='Custom') def reverse_geocoding(lat, lon, feature, parent): try: location = locator.reverse(Point(lat, lon)) return location.raw['display_name'] except: pass Here the Nominatim geocoder (free to choose) from the GeoPy geocoding Python library was used, for more details, please check [the documentation](https://geopy.readthedocs.io/en/latest/#nominatim). Also a point Mind that additionally installation of `geopy` library inside the QGIS is required, the *how-to process* described in [Installing 3rd party python libraries for QGIS on Windows](https://gis.stackexchange.com/questions/141320/installing-3rd-party-python-libraries-for-qgis-on-windows), [Installing Python packages in QGIS 3 (for Windows)](https://landscapearchaeology.org/2018/installing-python-packages-in-qgis-3-for-windows/), and [Install QGIS through OSGeo4W](https://www.giscourse.com/install-qgis-through-osgeo4w/) (also applicable for Python packages). And afterwards, I ran the model with [![run][4]][4] (<kbd>F5</kbd> as shortcut) as got the output [![result][5]][5] ---------- Another solution that I may inspire with, described in [the @csk's answer](https://gis.stackexchange.com/a/300619/99589). So, you may try to find and download the reference point data with addresses for your area, and then simply using the ["Join attributes by nearest"](https://docs.qgis.org/3.16/en/docs/user_manual/processing_algs/qgis/vectorgeneral.html#join-attributes-by-nearest) geoalgorithm in the Graphical Modeler. [1]: https://i.sstatic.net/iJBsf.png [2]: https://i.sstatic.net/lJo3Y.png [3]: https://i.sstatic.net/5jdYZ.png [4]: https://i.sstatic.net/vHkKx.png [5]: https://i.sstatic.net/wIPoR.png