# neuralgcm/neuralgcm

Hybrid ML + physics model of the Earth's atmosphere

Repository: https://github.com/neuralgcm/neuralgcm
Canonical: https://ross.abutalabs.com/products/neuralgcm
Homepage: https://neuralgcm.readthedocs.io
Language: Python
License: Apache-2.0
License Family: permissive
Last push: 2026-07-14T16:05:06+00:00

## Health v2 (maintenance only)
Score: 69/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 92, release rhythm 41, longevity 67
- inputs: {"age_days": 940, "days_push": 50, "days_rel": 398, "gap_med": 19, "n_releases_24m": 4}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1008, forks 133 (observed 2026-08-28T04:03:12.547166+00:00)

## What it is
NeuralGCM is a Python library for building hybrid machine-learning/physics models of the Earth's atmosphere for weather and climate simulation. It combines learned neural components with traditional general circulation model dynamics.

## Use cases
- simulate weather with a hybrid ML physics model
- run atmospheric climate simulations in Python
- forecast weather using neural general circulation models
- research machine learning for weather and climate modeling
- build hybrid atmospheric models combining ML and physics

## When to choose
- you need state-of-the-art hybrid ML/physics atmospheric simulation
- you are doing research on ML-based weather or climate modeling
- you want a Python library with pretrained model weights for atmospheric simulation

## When to avoid
- you need a simple point-and-click weather forecast app
- you lack the computational resources (GPUs/TPUs) for atmospheric simulation
- you need operational meteorology tooling outside Python

## Facets
- artifact type: library
- maturity: active
- function: machine-learning, simulation, data-science
- domain: machine-learning, data-science
- platform: python
- tags: weather, climate-modeling, atmospheric-science, hybrid-ml-physics, numerical-weather-prediction, jax, algorithms, linux, macos

## Member repositories
- neuralgcm/neuralgcm (main) score 69

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:12.547166+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 2026-08-30T07:12:12.312945+00:00, confidence not recorded.
  - readme: https://github.com/neuralgcm/neuralgcm (fetched 2026-08-28T04:03:12.547166+00:00, sha 0e0c6f4e585a)
  - registry_pypi: https://pypi.org/pypi/neuralgcm/json (fetched 2026-08-29T13:13:02.701243+00:00, sha 1a5d7ff78696)
- Data as of 2026-08-30T08:39:29.467469+00:00.