# ACE: A fast, skillful learned global atmospheric model for climate prediction

*Source* · `src-wattmeyer-2023-ace` · status: **n/a**

| field | value |
|---|---|
| kind | Preprint |
| venue | arXiv |
| date | 2023-10-03 |
| url | https://arxiv.org/abs/2310.02074 |
| arxiv | 2310.02074 |
| accessed | 2026-11-01 |
| notes | A 200M-parameter learned emulator of a 100-km global atmospheric model, run stably for long climate rollouts. |

## Authors

- Oliver Watt-Meyer
- Gideon Dresdner
- Jeremy McGibbon
- Spencer K. Clark
- Brian Henn
- James Duncan
- Noah D. Brenowitz
- Karthik Kashinath
- Michael S. Pritchard
- Boris Bonev
- Matthew E. Peters
- Christopher S. Bretherton

## Backlinks (computed)

**cited by:** [Learned regional weather](../opportunity/learned-weather-layer.md)

## Citation

Oliver Watt-Meyer, Gideon Dresdner, Jeremy McGibbon, Spencer K. Clark, Brian Henn, James Duncan, Noah D. Brenowitz, Karthik Kashinath, Michael S. Pritchard, Boris Bonev, Matthew E. Peters, Christopher S. Bretherton (2023-10-03). *ACE: A fast, skillful learned global atmospheric model for climate prediction*. arXiv https://arxiv.org/abs/2310.02074

---
Provenance: created in run `2026-11-01-es-v3-prior-art`.