Herbie is a Python library for downloading and reading numerical weather prediction (NWP) model data in GRIB2 format. It provides a unified interface to access high-resolution weather data from sources like HRRR, RAP, GFS, and others. The current version is 2026.3.0, with a regular release cadence (monthly).
pip install herbie-dataNo compatibility data collected yet for this library.
Verified import paths — ran on the pinned version, not inferred.
This example downloads the 2-meter temperature field from the HRRR model for January 1, 2022.
Set the environment variable HERBIE_DATA_DIR before creating a Herbie object.
Use the .xarray() method with the field name as a string.
Pass the correct product string, e.g., product='pgrb2b.0p25' for GFS 0.25 deg.
pip install herbie-data and then use 'from herbie import Herbie'.
List available fields with H.inventory() and check the exact parameter name.
Use the exact model name: model='GFS'.