ai-models (ECMWF)
ECMWF's unified framework and command-line tool to run AI-based weather forecasting models (GraphCast, Aurora, Pangu, NeuralGCM, FourCastNet) with operational ECMWF data infrastructure, enabling standardized inference and benchmarking across state-of-the-art meteorological AI systems (ECMWF, 576+ stars)
README
ai-models DISCLAIMER \[!IMPORTANT\] This software is Archived and subject to ECMWF's guidelines on Software Maturity. This project is no longer actively maintained or developed. It will remain available for reference or historical purposes, but should NOT be used for any active operational purposes. The ai-models command is used to run AI-based weather forecasting models. These models need to be installed independently. Usage Although the source code ai-models and its plugins are available…
- Repository
- github.com/ecmwf-lab/ai-models
Source attribution
- GitHub — github.com/ecmwf-lab/ai-models
- Awesome AI for Science — github.com/ecmwf-lab/ai-models
Related resources
Shanghai AI Lab's deep learning-based global weather forecasting model pushing skillful forecasts beyond 10 days lead, with open-source inference code and pretrained ONNX model weights (arXiv 2023)
Microsoft's AI-powered geospatial Earth science application for natural-language exploration, visualization, and analysis of 130+ satellite collections, with STAC integration, multi-agent backend, MCP server, and deployable React/FastAPI stack (MIT, 2025)
First foundation model for weather and climate by Microsoft, Vision Transformer-based architecture trained on heterogeneous datasets (ICML 2023)
Google Research's hybrid ML/physics atmospheric model combining learned dynamics with physical constraints, outperforming traditional models on 2-15 day forecasts and 40-year climate simulation, developed with ECMWF (Nature 2024)
World's first fully open, accelerated weather AI software stack with Medium Range forecasting and Nowcasting models using generative AI (January 2026)
Microsoft's foundation model for the Earth system supporting weather, air pollution, and ocean wave forecasting at multiple resolutions, trained on 1M+ hours of diverse atmospheric data (Nature 2025)