ORBIT-2 Weather and Climate Downscaling Software Repository
- Oak Ridge National Laboratory
- AMD
- Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (United States)
ORBIT-2 is a scalable foundation model for global, hyper-resolution climate and weather downscaling. ORBIT-2 incorporates two key innovations: (1) Residual Slim ViT (Reslim), a lightweight architecture with residual learning and Bayesian regularization for efficient, robust prediction; and (2) TILES, a tile-wise sequence scaling algorithm that reduces self-attention complexity from quadratic to linear, enabling long-sequence processing and massive parallelism. ORBIT-2 scales to 10 billion parameters across 65,536 GPUs, achieving up to 4.1 ExaFLOPS sustained throughput and 74–98% strong scaling efficiency. It supports downscaling to 0.9 km global resolution and processes sequences up to 4.2 billion tokens. On 7 km resolution benchmarks, ORBIT-2 achieves high accuracy with 𝑅2 scores in range of 0.98–0.99 against observation data.
- Short Name / Acronym:
- ORBIT-2
- Software Type:
- Scientific
- License(s):
- MIT License
- Research Organization:
- Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (United States)
- Sponsoring Organization:
- USDOE Office of Science (SC)Primary Award/Contract Number:AC05-00OR22725
- DOE Contract Number:
- AC05-00OR22725
- Code ID:
- 167711
- OSTI ID:
- code-167711
- Country of Origin:
- United States
Similar Records
Downscaling Extended Weather Forecasts for Hydrologic Prediction
ORBIT-2 Dataset for Scaling Exascale Vision Foundation Models for Weather and Climate Downscaling