Pith. sign in

REVIEW 26 cited by

GenCast: Diffusion-based ensemble forecasting for medium-range weather

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2312.15796 v2 pith:VEKGJNIL submitted 2023-12-25 cs.LG physics.ao-ph

classification cs.LGphysics.ao-ph
keywords weathergencastensembleforecastsgreatermedium-rangedecisionsforecast
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Weather forecasts are fundamentally uncertain, so predicting the range of probable weather scenarios is crucial for important decisions, from warning the public about hazardous weather, to planning renewable energy use. Here, we introduce GenCast, a probabilistic weather model with greater skill and speed than the top operational medium-range weather forecast in the world, the European Centre for Medium-Range Forecasts (ECMWF)'s ensemble forecast, ENS. Unlike traditional approaches, which are based on numerical weather prediction (NWP), GenCast is a machine learning weather prediction (MLWP) method, trained on decades of reanalysis data. GenCast generates an ensemble of stochastic 15-day global forecasts, at 12-hour steps and 0.25 degree latitude-longitude resolution, for over 80 surface and atmospheric variables, in 8 minutes. It has greater skill than ENS on 97.4% of 1320 targets we evaluated, and better predicts extreme weather, tropical cyclones, and wind power production. This work helps open the next chapter in operational weather forecasting, where critical weather-dependent decisions are made with greater accuracy and efficiency.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 26 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. OpenAlex reports about 62 citations worldwide. Full citation record

  1. Timestep-Conditioned Transformers for Global Weather Forecasting

    cs.LG 2026-08 conditional novelty 7.0 of 10

    A single weather transformer with inference-time timestep conditioning matches specialist models across 1 to 24 hour steps and improves rollout stability under mixed-timestep training.

  2. Round-Trip Consistency: Bidirectional Diffusion Models Can Predict Their Own Rollout Errors

    stat.ML 2026-08 conditional novelty 7.0 of 10

    A model's forward-then-backward round-trip error predicts, without ensembles or ground truth, how large its actual rollout error will be.

  3. Geospatial Diffusion-based Evolution Synthesis (GeoDES) for Storm-Centered Weather Augmentation

    cs.LG 2026-07 conditional novelty 7.0 of 10

    GeoDES generates realistic synthetic cyclone evolutions via 2D-pretrained, temporally-inflated diffusion with correlated noise, beating weather foundation models on storm-kinetics and energy-spectrum metrics.

  4. When is a System Discoverable from Data? Discovery Requires Chaos

    math.DS 2025-11 conditional novelty 7.0 of 10

    Uniquely identifying an ODE from trajectory data depends on the trajectory filling enough of the state space: chaos on a high-dimensional attractor yields analytic discoverability, while first integrals preclude it.

  5. The Effect of Stochasticity in Score-Based Diffusion Sampling: a KL Divergence Analysis

    cs.LG 2025-06 conditional novelty 7.0 of 10

    KL divergence bounds show stochasticity in diffusion sampling contracts error with exact scores, but for learned scores it can help or hurt depending on the time profile of the score error.

  6. Skillful joint probabilistic weather forecasting from marginals

    cs.LG 2025-06 conditional novelty 7.0 of 10

    FGN, a neural weather model trained only on per-location forecast scores, produces more accurate global ensemble forecasts than GenCast and captures realistic spatial correlations.

  7. HourGlass: A probabilistic data-driven temporal downscaler for global and regional weather forecasting

    physics.ao-ph 2026-07 accept novelty 6.5 of 10

    HourGlass probabilistically reconstructs hourly weather evolution between 6-hourly forecast states using CRPS training on NWP trajectories, preserving skill and small-scale variability better than deterministic downscalers.

  8. Prithvi-Precip: Integrating Satellite Observations into an Atmospheric AI Foundation Model for Precipitation Forecasting

    physics.ao-ph 2026-08 conditional novelty 6.0 of 10

    Training on IMERG satellite precipitation and directly ingesting satellite observations improves medium-range AI precipitation forecasts, with the largest gains at lead times under 40 hours.

  9. Less Tokens, Better Forecasts: Sparse Residual Routing for Efficient Weather Prediction

    cs.LG 2026-07 conditional novelty 6.0 of 10

    Sparse residual routing of 25% of spatial tokens through middle ViT blocks improves ERA5 weather forecast skill while cutting training time ~2.5x and peak memory >2.2x.

  10. Flow Learners for PDEs: Toward a Physics-to-Physics Paradigm for Scientific Computing

    cs.LG 2026-04 unverdicted novelty 6.0 of 10

    Learned PDE solving should target transport over admissible futures via flow learners, not snapshot state regression.

  11. Scaling Laws of Global Weather Models

    cs.LG 2026-02 conditional novelty 6.0 of 10

    Across five global weather models, validation loss follows power-law scaling, with wider architectures and larger training datasets outperforming deeper or smaller-data configurations.

  12. MoWE : A Mixture of Weather Experts

    cs.LG 2025-09 conditional novelty 6.0 of 10

    MoWE, a ViT-based gating network, combines forecasts from Pangu, Aurora, and FCN3 with per-grid-point weights and beats each expert and the simple mean in RMSE.

  13. IndiaWeatherBench: A Dataset and Benchmark for Data-Driven Regional Weather Forecasting over India

    cs.LG 2025-08 conditional novelty 6.0 of 10

    A curated 12-km, 43-channel, 2000-2019 regional reanalysis dataset with standardized splits and UNet/Stormer/GraphCast/Hi baselines for data-driven weather forecasting over India.

  14. Projecting U.S. coastal storm surge risks and impacts with deep learning

    physics.ao-ph 2025-06 conditional novelty 6.0 of 10

    Running a neural-network surge emulator over 900,000 synthetic hurricanes, the authors project that the US population at risk from the 100-year storm surge flood rises about 50% by 2100 under SSP5-8.5, driven mainly b...

  15. Fusion of multi-source precipitation records via coordinate-based generative model

    physics.ao-ph 2025-06 conditional novelty 6.0 of 10

    A coordinate-based diffusion model fuses multi-source precipitation records and corrects biases in unseen operational forecasts.

  16. LaDCast: A Latent Diffusion Model for Medium-Range Ensemble Weather Forecasting

    cs.LG 2025-06 conditional novelty 6.0 of 10

    A latent diffusion model generates medium-range global weather ensembles at 1.5 degrees that match ECMWF IFS-ENS deterministic skill at lower compute, with weaker probabilistic spread and anecdotal cyclone advantages.

  17. DEF: Diffusion-augmented Ensemble Forecasting

    cs.LG 2025-06 conditional novelty 6.0 of 10

    A conditional diffusion model that generates perturbed initial states can turn any deterministic neural weather forecast model into an ensemble, with measured error reduction on a single ERA5 case study.

  18. Hierarchical Implicit Neural Emulators

    cs.LG 2025-06 conditional novelty 6.0 of 10

    Feeding a hierarchy of predicted coarse-grained future states into an autoregressive neural emulator greatly improves long-term stability for 2D turbulent flow forecasting.

  19. FLEX: A Backbone for Diffusion-Based Modeling of Spatio-temporal Physical Systems

    cs.LG 2025-05 conditional novelty 6.0 of 10

    A hybrid U-Net and Transformer diffusion backbone trained in residual space with a velocity parametrization outperforms baselines on 2D turbulence super-resolution and forecasting, and generalizes zero-shot to unseen ...

  20. Improving the Predictability of the Madden-Julian Oscillation at Subseasonal Scales with Gaussian Process Models

    math.NA 2025-05 reject novelty 6.0 of 10

    A Gaussian process MJO forecaster with empirical correlations claims better skill than an ANN for the first five lead days and probabilistic coverage extending beyond three weeks after a covariance correction.

  21. CSU-PCAST: A Dual-Branch Transformer Framework for medium-range ensemble Precipitation Forecasting

    physics.ao-ph 2025-10 conditional novelty 5.0 of 10

    A transformer-based ensemble weather model trained on ERA5 with IMERG precipitation labels beats GEFS precipitation skill at short-to-medium lead times in January and July 2023.

  22. A Geometry-Aware AI Emulator for the Coupled Whole Atmosphere from Earth Surface to the Ionosphere and Thermosphere

    physics.space-ph 2025-06 reject novelty 5.0 of 10

    CAM-NET, an SFNO-based neural emulator, is claimed to reproduce WACCM-X whole-atmosphere variability at over 1000x speedup, but the paper provides no quantitative validation, code, or data to support the claim.

  23. Arnoldi Singular Vector perturbations for machine learning weather prediction

    physics.ao-ph 2025-06 conditional novelty 5.0 of 10

    An adjoint-free Arnoldi Singular Vector method applied to the Pangu Weather ML model generates flow-dependent initial perturbations that grow immediately, unlike damped random noise.

  24. A multi-scale loss formulation for learning a probabilistic model with proper score optimisation

    physics.ao-ph 2025-06 conditional novelty 5.0 of 10

    Adding a multi-scale loss to proper-score-trained AIFS-CRPS reduces small-scale variability in forecasts without changing skill scores.

  25. Physics-Guided Learning of Meteorological Dynamics for Weather Downscaling and Forecasting

    cs.LG 2025-05 conditional novelty 5.0 of 10

    PhyDL-NWP trains neural surrogates with a fitted PDE regularizer and a latent force term, improving weather downscaling and fine-tuning forecasts over 17 baselines.

  26. Machine learning for modelling unstructured grid data in computational physics: a review

    cs.LG 2025-02 conditional novelty 2.0 of 10

    A broad review of machine learning techniques for modeling unstructured mesh data in computational physics, with a taxonomy, a qualitative comparison, and a list of public benchmarks.

Pith tools