NVIDIA Earth2Studio
Next-generation AI weather and climate modeling.
Install | User Guide | Examples | API | Skills
Access a leading collection of weather and climate AI models, production-ready data sources, composable inference APIs, and GPU-accelerated workflows in one Python package.
Get Started
Paste this into your coding agent (Claude, Codex, OpenCode, Cursor, etc.):
Help me get started with Earth2Studio.
https://github.com/NVIDIA/earth2studio
AGENTS.md routes usage and development tasks to the relevant
documentation and skills. Learn more, install, and start forecasting with just a few prompts:
# Learn about the latest release
Summarize new models and features in the latest Earth2Studio release.
Include the version, release date, and links to the release notes.
# Create a forecasting pipeline
Create a 10-day AI forecasting pipeline starting from the latest GFS analysis.
# Set up a Python environment for a specific model
Set up my environment for FourCastNet3 inference.
# Find a model for your forecast and hardware
Recommend the best model for a 2-week global forecast on one GPU.
# Explore models that predict a specific product
What models do you have that support predicting precipitation?
# Download weather data
Fetch ERA5 surface winds for March 2024 and save them to Zarr.
To keep the skills installed, add them from this repository:
npx skills add NVIDIA/earth2studio --skill earth2studio-install
npx skills add NVIDIA/earth2studio --skill earth2studio-discover
Installation and Examples
Each model has different dependencies; the install guide provides the exact commands for your setup.
| Resource | What's there |
|---|---|
| Install | Environments, model-specific extras, hardware |
| Examples | Deterministic, ensemble, diagnostic and coupled workflows |
| User Guide | Components, design philosophy and extending the package |
| API | Every prognostic, diagnostic and data source |
Main Features
- Forecast with global and regional prognostic AI models.
- Diagnose derived fields such as precipitation, wind speed and downscaled output.
- Assimilate sparse observations into global and regional analyses.
- Fetch analysis, forecast and observation data from public sources.
- Compose ensembles and coupled multi-model workflows.
- Store results to Zarr, NetCDF and more.
Documentation
Everything else (model zoo, data sources, news, tutorials) lives on the documentation site. See the changelog for what is new, CONTRIBUTING to get involved, and Earth-2 Open Models for NVIDIA's open weights and PhysicsNeMo for training recipes. For deploying workflows as an inference service, see PhysicsNeMo Serve.
License
Apache License 2.0, see the LICENSE file.
Metadata
Release files for earth2studio 1.0.0
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | Actions |
|---|---|---|---|---|
| earth2studio-1.0.0.tar.gz | 1.1 MB | Without provenance | Details for earth2studio-1.0.0.tar.gz |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| earth2studio-1.0.0-py3-none-any.whl | Python 3 | none | any | Details for earth2studio-1.0.0-py3-none-any.whl |
Total release size: 2.6 MB
Release files / earth2studio-1.0.0.tar.gz
| Download URL | earth2studio-1.0.0.tar.gz |
|---|---|
| Size | 1.1 MB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
f84e9aaadf2c27a8831a15b10faf53c9cc2abfaad39b50dc0e33476cbdd41bd4
|
|
BLAKE2b-256 checksum How to use checksums |
1bc31129a6821a6ce7e473e7de9263cb1f5774dabe4085f7c5d78fea498e1a6c
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/7.0.0 CPython/3.14.8
|
Release files / earth2studio-1.0.0-py3-none-any.whl
| Download URL | earth2studio-1.0.0-py3-none-any.whl |
|---|---|
| Size | 1.5 MB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
37720b60f097916582545c7590e338f8da074c9a9d16177ba6a0ae4194572e3a
|
|
BLAKE2b-256 checksum How to use checksums |
b5909d3197e72d9c966c0663f531cc5ff460376c54acfa6acf9e61e07d0d451f
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/7.0.0 CPython/3.14.8
|
Security for earth2studio 1.0.0
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
No provenance found
None of the files in this release were uploaded with provenance metadata.
A note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.