GEMv3
The latest in AI weather models, with reliable guidance from day 1 to day 126.
GEMv3
The latest in AI weather models, with reliable guidance from day 1 to day 126.
Output variables cover key surface fields and atmospheric dynamics. See documentation for details.
2 m instantaneous; timestep min, max, and mean.
Mean precipitation rate and snowfall rate.
10 m and 100 m mean wind speed; 10 m max gusts; 100m u/v instantaneous components.
Surface downward shortwave flux; total cloud cover.
Mean sea-level pressure; 500 hPa geopotential.
ENSO, MJO, QBO, and ZMZW indices.
15-day benchmark CRPS scores against ECMWF ENS, ECMWF AIFS, and Google WeatherNext. Scores computed over July 2025 to March 2026. GEMv3 outperforms ECMWF ENS across all variables, is competitive with other state-of-the-art AI models on large-scale dynamics, and achieves best in class performance on 2m temperature, particularly the sub-timestep aggregations which GEMv3 outputs natively.
pip install "xarray>=2025.6.0" "zarr>=3" "pcodec>=1" s3fsimport xarray as xr
ds = xr.open_zarr(
"https://gemv3-reforecast.salient-open-data.com/forecast",
chunks=None,
)
t2m = ds.sel(
forecast_date="2025-01-01",
lat=42.36, lon=-71.06, method="nearest",
)["2m_temperature"]
t2m = t2m.assign_coords(valid_time=t2m.forecast_date + t2m.lead
).swap_dims(lead="valid_time")
t2m.plot(hue="sample", add_legend=False)See docs for additional details.
Publicly accessible, for verification and backtesting.
Live forecasts from January 2026 onward.