Weather4cast 2025

The Weather4cast competition presented at NeurIPS 2025 focuses on topics of high impact and practical value for our society – predicting future weather and changes of our environment. Unusual weather is increasing all over the world, reflecting ongoing climate change, and affecting communities in agriculture, transport, public health and safety, etc.

Can you predict future rain patterns with modern machine learning algorithms? Apply spatio-temporal modelling to complex dynamic systems! Get access to unique large-scale data and demonstrate temporal and spatial transfer learning under strong distributional shifts! We provide a super-resolution challenge of high relevance to forecasting unusual local events, where you need to predict future weather as measured by ground-based hi-res rain radar weather stations. Exploit data fusion of high-resolution rain radar maps movies combined, large-scale multi-band satellite images, and static information like topology!

Winning models will advance key areas of methods research in machine learning, of relevance beyond the application domain. The authors of competitive submissions will be invited to present at the NeurIPS Competition Track (virtual) and contribute to a joint article in a special PMLR NeurIPS Competition Proceedings volume.

This year, on the road to flexible foundation models, we move from basic precipitation prediction to testing generalization performance and emergent capabilities of probabilistic models on a set of downstream tasks, such as cumulative rainfall and unusual event prediction.

You can already get going with the data provided. In the next months, we aim to also provide a stronger baseline model and we’re also working on a special pollution prediction task.

Sign up to the forums to stay in the loop!