Data

Tools

Indexes

Applications

Experiments

Open Sustainable Technology

CREDIT

An open software platform to train and deploy AI atmospheric prediction model.
https://github.com/ncar/miles-credit

Category: Atmosphere
Sub Category: Meteorological Observation and Forecast

Keywords from Contributors

weather grib hugo

Last synced: about 12 hours ago
JSON representation

Repository metadata

NCAR MILES Community Research Earth Digital Intelligence Twin (CREDIT): research platform for ML Earth system prediction models.

README.md

NSF NCAR MILES Community Research Earth Digital Intelligence Twin (CREDIT)

DOI

PyPI

CREDIT npj Climate and Atmospheric Science Article

About

CREDIT is an open software platform to train and deploy AI atmospheric prediction models. CREDIT offers fast models
that can be flexibly configured both in terms of input data and neural network architecture. The interface is designed
to be user-friendly and enable fast spin-up and iteration. CREDIT is backed by the AI and atmospheric science expertise
of the MILES group and the NSF National Center for Atmospheric Research, leading to design choices that balance advanced
AI/ML with our physical knowledge of the atmosphere.

CREDIT has reached its first stable release with a full set of models, training, and deployment options. It continues
to be under active development. Please contact the MILES group if you have any questions about CREDIT.

New to CREDIT? See QUICKSTART.md to get up and running locally, or the full online Quickstart for detailed guidance.

AI assistant

credit ask is a unified AI assistant — it automatically runs in agent mode (reads your
files, runs shell commands, iterates until it has a confident answer) when an Anthropic key
is available, or falls back to simple chat (Groq, OpenRouter, Gemini, or OpenAI) otherwise.

pip install "miles-credit[ask]"

# Free — no card needed (pick either)
export GROQ_API_KEY=gsk_...          # https://console.groq.com
export OPENROUTER_API_KEY=sk-or-...  # https://openrouter.ai

# Or use your own Anthropic key for full agent mode (~$0.01–0.05/session)
export ANTHROPIC_API_KEY=sk-ant-...  # https://console.anthropic.com

credit ask "why is my loss not decreasing?"
credit ask -c my_run.yml "why did my training run crash?"

The assistant picks the first provider with an available API key. Supported keys:

Env var Provider / model Mode
ANTHROPIC_API_KEY Anthropic (Claude) full agent mode
GROQ_API_KEY Groq (Llama 3, free) simple chat
OPENROUTER_API_KEY OpenRouter (Qwen3-Next-80B, free) simple chat
GOOGLE_API_KEY Google (Gemini) simple chat
OPENAI_API_KEY OpenAI (GPT-4o) simple chat

See the AI Assistant documentation for the full reference.

MILES CREDIT also provides more detailed documentation with installation
instructions, how to get started training and deploying models, how to interpret the config files, and full API docs.

Citing CREDIT

If you are interested in using CREDIT as part of your research, please cite the following paper:
Schreck, J.S., Sha, Y., Chapman, W. et al. Community Research Earth Digital Intelligence Twin: a scalable framework
for AI-driven Earth System Modeling. npj Clim Atmos Sci 8, 239 (2025). https://doi.org/10.1038/s41612-025-01125-6

Model Weights and Data

Model weights for the CREDIT 6-hour WXFormer and FuXi models and the 1-hour WXFormer are available on huggingface.

Processed ERA5 Zarr Data are available for download through Globus (requires free account) through the CREDIT ERA5 Zarr Files collection.

Scaling/transform values for normalizing the data are available through Globus here.

CREDIT also supports realtime runs generated from deterministic Google Cloud GFS files
and raw cube sphere GEFS files.

Support

This software is based upon work supported by the NSF National Center for Atmospheric Research, a major facility sponsored by the
U.S. National Science Foundation under Cooperative Agreement No. 1852977 and managed by the University Corporation for Atmospheric Research. Any opinions, findings and conclusions or recommendations
expressed in this material do not necessarily reflect the views of NSF. Additional support for development was provided by
The NSF AI Institute for Research on Trustworthy AI for Weather, Climate, and Coastal Oceanography (AI2ES) with grant
number RISE-2019758 and by Schmidt Sciences, LLC.

Owner metadata

GitHub Events

Total
Last Year

Committers metadata

Last synced: 3 days ago

Total Commits: 2,293
Total Committers: 28
Avg Commits per committer: 81.893
Development Distribution Score (DDS): 0.756

Commits in past year: 1,014
Committers in past year: 21
Avg Commits per committer in past year: 48.286
Development Distribution Score (DDS) in past year: 0.679

Name Email Commits
John Schreck j****k@g****m 559
Yingkai Sha y****a@g****m 511
David John Gagne d****e@g****m 382
Seth McGinnis m****s@u****u 148
charlie-becker c****r@u****u 137
kevinyang-cky k****g@u****u 106
willychap w****n@u****u 85
danineamati d****i@g****m 74
dkimpara d****a@u****u 51
dkimpara d****a@g****m 50
Katelyn FitzGerald 7****d 47
Negin Sobhani n****3@g****m 33
Arnold Kazadi a****o@g****m 23
dependabot[bot] 4****] 13
sgpearse p****e@u****u 13
ggantos g****s@u****u 13
pre-commit-ci[bot] 6****] 12
Ben Kirk b****k@u****u 8
Stanley Akor s****r@c****u 6
Stanley Akor s****r@c****u 6
Stanley Akor s****r@c****u 4
Nihanth Wagmi Cherukuru n****u@u****u 3
Jay Rothenberger 3****r 3
Stanley Akor s****r@u****u 2
Sai Asish Y s****5@g****m 1
“kjmayer” “****r@u****” 1
Stanley Akor s****r@c****u 1
unknown d****8@g****m 1

Committer domains:

Issue and Pull Request metadata

Last synced: 13 days ago

Total issues: 56
Total pull requests: 228
Average time to close issues: 4 months
Average time to close pull requests: 9 days
Total issue authors: 15
Total pull request authors: 20
Average comments per issue: 0.98
Average comments per pull request: 0.82
Merged pull request: 170
Bot issues: 0
Bot pull requests: 5

Past year issues: 38
Past year pull requests: 69
Past year average time to close issues: about 1 month
Past year average time to close pull requests: 9 days
Past year issue authors: 12
Past year pull request authors: 16
Past year average comments per issue: 0.82
Past year average comments per pull request: 1.41
Past year merged pull request: 34
Past year bot issues: 0
Past year bot pull requests: 5

More stats: https://issues.ecosyste.ms/repositories/lookup?url=https://github.com/ncar/miles-credit

Top Issue Authors

  • charlie-becker (11)
  • djgagne (11)
  • kafitzgerald (8)
  • dkimpara (8)
  • yingkaisha (4)
  • NihanthCW (2)
  • sethmcg (2)
  • kevinyang-cky (2)
  • jsschreck (2)
  • kjmayer (1)
  • jshaw35 (1)
  • tmerlis (1)
  • AidanJanney (1)
  • WillyChap (1)
  • sgpearse (1)

Top Pull Request Authors

  • jsschreck (71)
  • yingkaisha (33)
  • djgagne (32)
  • dkimpara (18)
  • WillyChap (14)
  • kanz76 (11)
  • charlie-becker (10)
  • kafitzgerald (9)
  • sethmcg (5)
  • dependabot[bot] (4)
  • stanleyakor1 (4)
  • sgpearse (3)
  • NihanthCW (3)
  • benkirk (2)
  • kjmayer (2)

Top Issue Labels

  • bug (5)
  • enhancement (2)
  • help wanted (1)
  • github_actions (1)

Top Pull Request Labels

  • github_actions (4)
  • dependencies (4)
  • documentation (1)
  • enhancement (1)
  • bug (1)

Package metadata

proxy.golang.org: github.com/ncar/miles-credit

  • Homepage:
  • Documentation: https://pkg.go.dev/github.com/ncar/miles-credit#section-documentation
  • Licenses: apache-2.0
  • Latest release: v2026.1.1+incompatible (published 2 months ago)
  • Last Synced: 2026-06-24T09:42:32.173Z (12 days ago)
  • Versions: 5
  • Dependent Packages: 0
  • Dependent Repositories: 0
  • Rankings:
    • Dependent packages count: 5.395%
    • Average: 5.576%
    • Dependent repos count: 5.758%
proxy.golang.org: github.com/NCAR/miles-credit

  • Homepage:
  • Documentation: https://pkg.go.dev/github.com/NCAR/miles-credit#section-documentation
  • Licenses: apache-2.0
  • Latest release: v2026.1.1+incompatible (published 2 months ago)
  • Last Synced: 2026-06-24T09:42:32.553Z (12 days ago)
  • Versions: 5
  • Dependent Packages: 0
  • Dependent Repositories: 0
  • Rankings:
    • Dependent packages count: 5.395%
    • Average: 5.576%
    • Dependent repos count: 5.758%

Dependencies

.github/workflows/python-package-conda.yml actions
  • actions/checkout v4 composite
  • mamba-org/setup-micromamba v1 composite
.github/workflows/python-publish.yml actions
  • actions/checkout v4 composite
  • actions/download-artifact v4 composite
  • actions/setup-python v5 composite
  • actions/upload-artifact v4 composite
  • pypa/gh-action-pypi-publish release/v1 composite
pyproject.toml pypi
  • bridgescaler *
  • cartopy *
  • dask *
  • dask-jobqueue *
  • distributed *
  • echo-opt *
  • einops *
  • fsspec *
  • gcsfs *
  • haversine *
  • matplotlib *
  • myst_parser *
  • netcdf4 *
  • numpy *
  • pandas *
  • pre-commit *
  • pvlib *
  • pyarrow *
  • pysteps *
  • pytest *
  • pyyaml *
  • ruff *
  • scikit-learn *
  • segmentation-models-pytorch *
  • sphinx *
  • sphinx-autoapi *
  • sphinx-book-theme *
  • torch *
  • torch-harmonics *
  • torch_geometric *
  • torchvision *
  • xarray *
  • zarr *
requirements.txt pypi
  • bridgescaler *
  • cartopy *
  • dask *
  • dask-jobqueue *
  • distributed *
  • echo-opt *
  • einops *
  • fsspec *
  • gcsfs *
  • haversine *
  • jupyter *
  • matplotlib *
  • netcdf4 *
  • numpy *
  • pandas *
  • pre-commit *
  • pvlib *
  • pyarrow *
  • pytest *
  • pyyaml *
  • ruff *
  • scikit-learn *
  • segmentation-models-pytorch *
  • torch *
  • torch-harmonics *
  • torch_geometric *
  • torchvision *
  • xarray *
  • zarr *
setup.py pypi

Score: -Infinity