# google-deepmind/weathernext

Repository: https://github.com/google-deepmind/weathernext
Canonical: https://ross.abutalabs.com/products/weathernext
Language: Python
License: Apache-2.0
License Family: permissive
Topics: weather, weather-forecast
Last push: 2026-08-11T12:29:00+00:00

## Health v2 (maintenance only)
Score: 82/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 97, release rhythm 64, longevity 81
- inputs: {"age_days": 1146, "days_push": 22, "days_rel": 27, "gap_med": 332.5, "n_releases_24m": 3}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 7596, forks 973 (observed 2026-08-28T04:10:01.862099+00:00)

## What it is
Google DeepMind's WeatherNext family of AI weather forecasting models, including WeatherNext 2, GraphCast, and GenCast, with code and pretrained weights to run global medium-range atmospheric and cyclone forecasts. It also points to daily forecast data feeds on Google Cloud, WeatherLab, and OpenMeteo.

## Use cases
- run AI-based global weather forecasts
- forecast tropical cyclone tracks with machine learning
- generate ensemble weather predictions with diffusion models
- run GraphCast or GenCast locally on pretrained weights
- access daily WeatherNext forecast data feeds via API

## When to choose
- you want to run state-of-the-art medium-range weather models yourself
- you need cyclone track forecasting with pretrained AI models
- you are doing research on ML-based weather prediction

## When to avoid
- you only need forecast data without running models - use the hosted data feeds instead
- you need local high-resolution or nowcasting models
- you lack GPU/TPU resources, as inference is compute-intensive

## Facets
- artifact type: library
- maturity: active
- function: machine-learning, deep-learning, simulation, data-science
- domain: weather, machine-learning, artificial-intelligence, data-science
- platform: python, cloud
- tags: weather-forecasting, graph-neural-networks, diffusion-models, cyclone-forecasting, pretrained-models, medium-range-forecasting, gpu

## Member repositories
- google-deepmind/weathernext (main) score 82

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:10:01.862099+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T17:36:52.325287+00:00, confidence not recorded.
  - readme: https://github.com/google-deepmind/weathernext (fetched 2026-08-28T04:10:01.862099+00:00, sha d96a2fe9833e)
- Data as of 2026-08-30T08:39:29.467469+00:00.