Skip to content

Archives

GenCast

  • GenCast

    Google DeepMind announce their new AI model for weather forecasting, in collaboration with the ECMWF:

    Today, in a paper published in Nature, we present GenCast, our new high resolution (0.25°) AI ensemble model. GenCast provides better forecasts of both day-to-day weather and extreme events than the top operational system, the European Centre for Medium-Range Weather Forecasts’ (ECMWF) ENS, up to 15 days in advance. We’ll be releasing our model’s code, weights, and forecasts, to support the wider weather forecasting community. […] GenCast is a diffusion model, the type of generative AI model that underpins the recent, rapid advances in image, video and music generation. However, GenCast differs from these, in that it’s adapted to the spherical geometry of the Earth, and learns to accurately generate the complex probability distribution of future weather scenarios when given the most recent state of the weather as input. To train GenCast, we provided it with four decades of historical weather data from ECMWF’s ERA5 archive. This data includes variables such as temperature, wind speed, and pressure at various altitudes. The model learned global weather patterns, at 0.25° resolution, directly from this processed weather data.
    It’s open source: https://github.com/google-deepmind/graphcast And here are the open-released model weights: https://console.cloud.google.com/storage/browser/dm_graphcast Graphcast (the previous iteration) has public forecasts published at https://charts.ecmwf.int/?query=GraphCast , under a CC-BY-NC-SA-4 licence — it would be great if the GenCast forecasts join this data set. Paper: https://arxiv.org/abs/2312.15796 This all looks really great, a fantastic commitment to (genuine) openness and open data, and the paper seems rigorous (to this amateur). Great stuff.

    (tags: forecasting weather ai gencast graphcast deepmind google ecmwf genai)