Skip to content
Back to search

earth2studio

Open-source deep-learning framework for exploring, building and deploying AI weather/climate workflows.

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Checked today

Check it deeper on your machine

It has changed since its scan on 11 Nov 2026.

As a Claude Code plugin, install it and then ask Claude Code:

/plugin marketplace add keithmackay/mackayi/plugin install scanskillsafety@mackayiDo a deep scan of https://github.com/NVIDIA/earth2studio with scanskillsafety

Or from a terminal, clone it and run it:

git clone https://github.com/keithmackay/scanskillsafety-skill ~/.claude/skills/scanskillsafetynode ~/.claude/skills/scanskillsafety/cli.cjs --deep 'https://github.com/NVIDIA/earth2studio'

It checks the exact commit: every file, its packages against OSV (api.osv.dev), how it compares with this page's scan (findsafeskills.com receives the repository's address), and, if you agree, Claude reading every file with your own Claude subscription. It never runs the code.

Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/nvidia--earth2studio/badge.svg)](/listings/nvidia--earth2studio)

HTML

<a href="/listings/nvidia--earth2studio"><img src="/api/listings/nvidia--earth2studio/badge.svg" alt="findsafeskills safety rating" /></a>

Install

i

Install instructions from the README

As written by the repo's author. findsafeskills has not verified these steps.

Running AI weather prediction can be done with just a few lines of code.

- For detailed installation steps, including model-specific installations, see the [install guide][e2studio_install_url]. - See the [examples][e2studio_examples_url] gallery providing different inference workflow samples. - Swap out [data sources][e2studio_data_api] or [models][e2studio_px_api] depending on your use case!

Tutorial

Agent-assisted setup

Automate setup with your preferred coding agent using NVIDIA Earth2Studio skills. Install the Earth2Studio skill set, then ask your favorite agent (Claude, Codex, OpenCode, etc) to recommend a model, configure an environment, or run a first deterministic forecast. Find more Earth2Studio skills in the [NVIDIA Skills catalog](https://build.nvidia.com/skills?q=earth2studio).

npx skills add NVIDIA/skills --skill earth2studio-install
npx skills add NVIDIA/skills --skill earth2studio-discover
npx skills add NVIDIA/skills --skill earth2studio-data-fetch
npx skills add NVIDIA/skills --skill earth2studio-deterministic-forecast

Example agent prompts:

Use the Earth2Studio discover skill to recommend a starter forecast workflow.
Use the Earth2Studio install skill to set up my environment for FourCastNet3 inference.
Create a script to fetch ERA5 surface winds data for March 2024.
Create a deterministic forecast workflow with GFS, FourCastNet3, and a Zarr output store.

NVIDIA FourCastNet3

from earth2studio.models.px import FCN3
from earth2studio.data import GFS
from earth2studio.io import ZarrBackend
from earth2studio.run import deterministic as run

model = FCN3.load_model(FCN3.load_default_package())
data = GFS()
io = ZarrBackend("outputs/fcn3_forecast.zarr")
run(["2025-01-01T00:00:00"], 10, model, data, io)

ECMWF AIFS

from earth2studio.models.px import AIFS
from earth2studio.data import IFS
from earth2studio.io import ZarrBackend
from earth2studio.run import deterministic as run

model = AIFS.load_model(AIFS.load_default_package())
data = IFS()
io = ZarrBackend("outputs/aifs_forecast.zarr")
run(["2025-01-01T00:00:00"], 10, model, data, io)

Google Graphcast

from earth2studio.models.px import GraphCastOperational
from earth2studio.data import GFS
from earth2studio.io import ZarrBackend
from earth2studio.run import deterministic as run

package = GraphCastOperational.load_default_package()
model = GraphCastOperational.load_model(package)
data = GFS()
io = ZarrBackend("outputs/graphcast_operational_forecast.zarr")
run(["2025-01-01T00:00:00"], 4, model, data, io)

> [!IMPORTANT] > Earth2Studio is an interface to third‑party models, checkpoints, and datasets. > Licenses for these assets are owned by their providers. > Ensure you have the rights to download, use, and (if applicable) redistribute each > model and dataset. > Links to the original license and source are often provided in the API docs for each > model/data source.

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

.claude-plugin/marketplace.json · Manifest · 37 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

This file has changed since it was scanned, so this is not exactly the text that was scanned, and no line is marked. The findings are listed with the safety scan above.

  1. {
  2. "name": "earth2studio",
  3. "description": "Earth2Studio skills for weather and climate AI workflows — model discovery, installation, data fetching, deterministic forecasting, and building new model/data source wrappers.",
  4. "version": "0.16.0",
  5. "owner": {
  6. "name": "NVIDIA"
  7. },
  8. "plugins": [
  9. {
  10. "name": "earth2studio",
  11. "source": {
  12. "source": "git-subdir",
  13. "url": "https://github.com/NVIDIA/earth2studio.git",
  14. "path": "./skills"
  15. },
  16. "description": "Earth2Studio skills for weather and climate AI — discover models, install the package, fetch data, run deterministic forecasts, and create new data source or prognostic model wrappers.",
  17. "version": "0.16.0",
  18. "author": {
  19. "name": "NVIDIA"
  20. },
  21. "homepage": "https://nvidia.github.io/earth2studio/",
  22. "repository": "https://github.com/NVIDIA/earth2studio",
  23. "license": "Apache-2.0",
  24. "category": "weather-climate",
  25. "tags": ["earth2studio", "weather", "climate", "AI", "forecasting", "NVIDIA"],
  26. "strict": false,
  27. "skills": [
  28. "./earth2studio-discover",
  29. "./earth2studio-install",
  30. "./earth2studio-data-fetch",
  31. "./earth2studio-deterministic-forecast",
  32. "./earth2studio-create-datasource",
  33. "./earth2studio-create-prognostic"
  34. ]
  35. }
  36. ]
  37. }
README.md · README · 229 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

This file has changed since it was scanned, so this is not exactly the text that was scanned, and no line is marked. The findings are listed with the safety scan above.

  1. <!-- markdownlint-disable MD002 MD033 MD041 MD053 -->
  2. > [!WARNING]
  3. > **Upcoming breaking release:** The next release after **0.20.0** will be **1.0.0**,
  4. > which will not be backward compatible. To keep existing workflows operational,
  5. > pin Earth2Studio to **0.20.0 or earlier** (`earth2studio<=0.20.0`) until you are ready to migrate.
  6. > See the [migration guide](https://github.com/NVIDIA/earth2studio/blob/1.0.0-rc/MIGRATION.md).
  7. <div align="center">
  8. # NVIDIA Earth2Studio
  9. [![python version][e2studio_python_img]][e2studio_python_url]
  10. [![license][e2studio_license_img]][e2studio_license_url]
  11. [![coverage][e2studio_cov_img]][e2studio_cov_url]
  12. [![mypy][e2studio_mypy_img]][e2studio_mypy_url]
  13. [![format][e2studio_format_img]][e2studio_format_url]
  14. [![ruff][e2studio_ruff_img]][e2studio_ruff_url]
  15. [![uv][e2studio_uv_img]][e2studio_uv_url]
  16. Earth2Studio is a Python-based package designed to get users up and running
  17. with AI Earth system models *fast*.
  18. Our mission is to enable everyone to build, research and explore AI driven weather and
  19. climate science.
  20. <!-- markdownlint-disable MD036 -->
  21. **- Earth2Studio Documentation -**
  22. <!-- markdownlint-enable MD036 -->
  23. [Install][e2studio_install_url] | [User-Guide][e2studio_userguide_url] |
  24. [Examples][e2studio_examples_url] | [API][e2studio_api_url]
  25. ![Earth2Studio README hero](https://huggingface.co/datasets/nvidia/earth2studio-assets/resolve/main/readme/v2/earth2studio-readme-hero.png?v2)
  26. </div>
  27. ## Quick start
  28. Running AI weather prediction can be done with just a few lines of code.
  29. - For detailed installation steps, including model-specific installations, see the
  30. [install guide][e2studio_install_url].
  31. - See the [examples][e2studio_examples_url] gallery providing different inference
  32. workflow samples.
  33. - Swap out [data sources][e2studio_data_api] or [models][e2studio_px_api] depending on
  34. your use case!
  35. ### Tutorial
  36. [![Earth2Studio Tutorial](https://huggingface.co/datasets/nvidia/earth2studio-assets/resolve/main/readme/v2/earth2studio-readme-quickstart-video.png?v1)](https://www.youtube.com/watch?v=Sog6aCapZeA)
  37. ### Agent-assisted setup
  38. Automate setup with your preferred coding agent using NVIDIA Earth2Studio skills.
  39. Install the Earth2Studio skill set, then ask your favorite agent (Claude, Codex, OpenCode, etc) to
  40. recommend a model, configure an environment, or run a first deterministic forecast.
  41. Find more Earth2Studio skills in the [NVIDIA Skills catalog](https://build.nvidia.com/skills?q=earth2studio).
  42. ![Earth2Studio agentic setup](https://huggingface.co/datasets/nvidia/earth2studio-assets/resolve/main/readme/v2/earth2studio-readme-agent-setup.png?v1)
  43. ```bash
  44. npx skills add NVIDIA/skills --skill earth2studio-install
  45. npx skills add NVIDIA/skills --skill earth2studio-discover
  46. npx skills add NVIDIA/skills --skill earth2studio-data-fetch
  47. npx skills add NVIDIA/skills --skill earth2studio-deterministic-forecast
  48. ```
  49. Example agent prompts:
  50. ```text
  51. Use the Earth2Studio discover skill to recommend a starter forecast workflow.
  52. Use the Earth2Studio install skill to set up my environment for FourCastNet3 inference.
  53. Create a script to fetch ERA5 surface winds data for March 2024.
  54. Create a deterministic forecast workflow with GFS, FourCastNet3, and a Zarr output store.
  55. ```
  56. ### NVIDIA FourCastNet3
  57. ```python
  58. from earth2studio.models.px import FCN3
  59. from earth2studio.data import GFS
  60. from earth2studio.io import ZarrBackend
  61. from earth2studio.run import deterministic as run
  62. model = FCN3.load_model(FCN3.load_default_package())
  63. data = GFS()
  64. io = ZarrBackend("outputs/fcn3_forecast.zarr")
  65. run(["2025-01-01T00:00:00"], 10, model, data, io)
  66. ```
  67. ### ECMWF AIFS
  68. ```python
  69. from earth2studio.models.px import AIFS
  70. from earth2studio.data import IFS
  71. from earth2studio.io import ZarrBackend
  72. from earth2studio.run import deterministic as run
  73. model = AIFS.load_model(AIFS.load_default_package())
  74. data = IFS()
  75. io = ZarrBackend("outputs/aifs_forecast.zarr")
  76. run(["2025-01-01T00:00:00"], 10, model, data, io)
  77. ```
  78. ### Google Graphcast
  79. ```python
  80. from earth2studio.models.px import GraphCastOperational
  81. from earth2studio.data import GFS
  82. from earth2studio.io import ZarrBackend
  83. from earth2studio.run import deterministic as run
  84. package = GraphCastOperational.load_default_package()
  85. model = GraphCastOperational.load_model(package)
  86. data = GFS()
  87. io = ZarrBackend("outputs/graphcast_operational_forecast.zarr")
  88. run(["2025-01-01T00:00:00"], 4, model, data, io)
  89. ```
  90. > [!IMPORTANT]
  91. > Earth2Studio is an interface to third‑party models, checkpoints, and datasets.
  92. > Licenses for these assets are owned by their providers.
  93. > Ensure you have the rights to download, use, and (if applicable) redistribute each
  94. > model and dataset.
  95. > Links to the original license and source are often provided in the API docs for each
  96. > model/data source.
  97. ## Latest News
  98. > [!NOTE]
  99. > Highlights from Earth2Studio 0.20.0.
  100. - [**HealDA v2**](https://nvidia.github.io/earth2studio/main/modules/generated/models/da/HealDAv2/),
  101. global data assimilation from 48 hours of satellite and conventional observations
  102. to a 0.25° atmospheric analysis.
  103. - [**StormScope Flash**](https://nvidia.github.io/earth2studio/main/modules/generated/models/px/StormScopeGOES/),
  104. accelerated GOES and MRMS nowcasting over CONUS, including crop-based inference.
  105. - [**Distilled CorrDiff ERA5-HRRR**](https://nvidia.github.io/earth2studio/main/modules/generated/models/dx/CorrDiffEra5Hrrr/),
  106. distilled generative downscaling from ERA5 to HRRR, now the default model.
  107. - [**InterpCRPSDiT**](https://nvidia.github.io/earth2studio/main/modules/generated/models/px/InterpCRPSDiT/),
  108. CRPS DiT temporal interpolation with sub-hourly forecast output.
  109. For a complete list of latest features and improvements see the [changelog](./CHANGELOG.md).
  110. ## Overview
  111. Earth2Studio is an *AI inference pipeline toolkit* focused on weather and climate
  112. applications that is designed to ride on top of different AI frameworks, model
  113. architectures, data sources and SciML tooling while providing a unified API.
  114. <div align="center">
  115. ![Earth2Studio model zoo](https://huggingface.co/datasets/nvidia/earth2studio-assets/resolve/main/readme/v2/earth2studio-readme-model-zoo.png?v3)
  116. ![Earth2Studio data sources](https://huggingface.co/datasets/nvidia/earth2studio-assets/resolve/main/readme/v2/earth2studio-readme-data-sources.png?v3)
  117. </div>
  118. The composability of the different core components in Earth2Studio easily allows the
  119. development and deployment of increasingly complex pipelines that may chain multiple
  120. data sources, AI models and other modules together.
  121. <div align="center">
  122. ![Earth2Studio composable pipelines](https://huggingface.co/datasets/nvidia/earth2studio-assets/resolve/main/readme/v2/earth2studio-readme-composability.png?v2)
  123. </div>
  124. The unified ecosystem of Earth2Studio provides users the opportunity to rapidly
  125. swap out components for alternatives.
  126. In addition to the largest model zoo of weather/climate AI models, Earth2Studio is
  127. packed with useful functionality such as optimized data access to cloud data stores,
  128. statistical operations and more to accelerate your pipelines.
  129. ### Earth-2 Open Models
  130. Access state of the art Nvidia open models for climate and weather: [Earth-2 Open Models](https://huggingface.co/collections/nvidia/earth-2).
  131. For training recipes for these models, see the [PhysicsNeMo repository][physicsnemo_repo_url].
  132. ## Contributors
  133. Check out the [contributing](CONTRIBUTING.md) document for details about the technical
  134. requirements and the user guide for higher level philosophy, structure, and design.
  135. ## License
  136. Earth2Studio is provided under the Apache License 2.0, refer to the
  137. [LICENSE file][e2studio_license_url] for full license text.
  138. <!-- Badge links -->
  139. [e2studio_python_img]: https://img.shields.io/badge/Python-3.11%20|%203.12%20|%203.13%20|%203.14-blue?style=flat-square&logo=python
  140. [e2studio_license_img]: https://img.shields.io/badge/License-Apache%202.0-green?style=flat-square
  141. [e2studio_format_img]: https://img.shields.io/badge/Code%20Style-Black-black?style=flat-square
  142. [e2studio_mypy_img]: https://img.shields.io/badge/mypy-Checked-blue?style=flat-square&labelColor=grey
  143. [e2studio_cov_img]: https://img.shields.io/codecov/c/github/nvidia/earth2studio?style=flat-square&logo=codecov
  144. [e2studio_ruff_img]: https://img.shields.io/endpoint?url=https://raw.githubusercontent.com/astral-sh/ruff/main/assets/badge/v2.json&style=flat-square
  145. [e2studio_uv_img]: https://img.shields.io/endpoint?url=https%3A%2F%2Fraw.githubusercontent.com%2Fastral-sh%2Fuv%2Fmain%2Fassets%2Fbadge%2Fv0.json&style=flat-square
  146. [e2studio_python_url]: https://www.python.org/downloads/
  147. [e2studio_license_url]: ./LICENSE
  148. [e2studio_format_url]: https://github.com/psf/black
  149. [e2studio_cov_url]: ./test/
  150. [e2studio_mypy_url]: https://mypy-lang.org/
  151. [e2studio_ruff_url]: https://github.com/astral-sh/ruff
  152. [e2studio_uv_url]: https://github.com/astral-sh/uv
  153. <!-- Doc links -->
  154. [e2studio_docs_url]: https://nvidia.github.io/earth2studio/
  155. [e2studio_install_url]: https://nvidia.github.io/earth2studio/main/userguide/about/install/
  156. [e2studio_userguide_url]: https://nvidia.github.io/earth2studio/main/userguide/
  157. [e2studio_examples_url]: https://nvidia.github.io/earth2studio/main/examples/
  158. [e2studio_api_url]: https://nvidia.github.io/earth2studio/main/modules/
  159. [e2studio_customization_url]: https://nvidia.github.io/earth2studio/main/examples/08_extend/
  160. [e2studio_px_url]: https://nvidia.github.io/earth2studio/main/userguide/components/prognostic/
  161. [e2studio_px_api]: https://nvidia.github.io/earth2studio/main/modules/models_px/
  162. [e2studio_dx_url]: https://nvidia.github.io/earth2studio/main/userguide/components/diagnostic/
  163. [e2studio_dx_api]: https://nvidia.github.io/earth2studio/main/modules/models_dx/
  164. [e2studio_data_url]: https://nvidia.github.io/earth2studio/main/userguide/components/datasources/
  165. [e2studio_data_api]: https://nvidia.github.io/earth2studio/main/modules/datasources_analysis/
  166. [e2studio_io_url]: https://nvidia.github.io/earth2studio/main/userguide/components/io/
  167. [e2studio_io_api]: https://nvidia.github.io/earth2studio/main/modules/io/
  168. [e2studio_pb_url]: https://nvidia.github.io/earth2studio/main/userguide/components/perturbation/
  169. [e2studio_pb_api]: https://nvidia.github.io/earth2studio/main/modules/perturbation/
  170. [e2studio_stat_url]: https://nvidia.github.io/earth2studio/main/userguide/components/statistics/
  171. [e2studio_stat_api]: https://nvidia.github.io/earth2studio/main/modules/statistics/
  172. [e2studio_lex_url]: https://nvidia.github.io/earth2studio/main/userguide/advanced/lexicon/
  173. <!-- Misc links -->
  174. [physicsnemo_repo_url]: https://github.com/NVIDIA/physicsnemo

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.