TL;DR
VIDRAFT opened two things on Hugging Face: Darwin-Prediction-Weather, a weather prediction system, and the Global Weather Model Leaderboard, a public board that scores the world's forecast models against what actually happened, every day. In the 2026 record, Darwin-Prediction-Weather has the lowest error in all four scored elements at every lead time from 1 to 7 days. The full write-up is on the Hugging Face blog.
The question: AI or physics?
For decades, forecasts came from physics models that solve the equations of the atmosphere on supercomputers. In the last three years, AI models trained on historical weather have joined them at the world's weather centres. Which one should you trust for tomorrow, for three days ahead, for a week ahead? Published papers answer that for past years. This board answers it for this year, updated daily, with the data open for anyone to check.
How the board scores
| Item | Rule |
|---|---|
| Models | ECMWF IFS and AIFS, NOAA GFS and AIGFS, DWD ICON, JMA GSM, ECCC GEM, MET Norway, Darwin-Prediction-Weather |
| Elements | Daily maximum temperature, daily minimum temperature, daily maximum wind speed, daily precipitation total |
| Where | 24 cities on six continents, up to 10 days ahead |
| Truth | ERA5 reanalysis, Copernicus Climate Change Service |
| Score | Root-mean-square error, lower is better |
| Fairness | Every model is scored on the same cases: days and cities where all listed models issued a forecast |
| Record | From 1 January 2026, updated once a day |
What the 2026 record shows
Two findings stand out.
AI models now lead. Among the models of the national and international weather centres, ECMWF's AI model AIFS has lower error than ECMWF's physics model IFS at every lead time, and NOAA's AIGFS has lower error than NOAA's GFS at every lead time.
Fusion goes further. Darwin-Prediction-Weather has the lowest error in all four elements, at every lead time from 1 to 7 days.
RMSE against ERA5, 24 cities, 2026 year to date (verified through 2026-11-07). The value in brackets is the next model, ECMWF AIFS in every cell.
| Element | 1 day | 3 days | 5 days | 7 days |
|---|---|---|---|---|
| Daily maximum temperature [°C] | 0.84 (1.35) | 1.17 (1.59) | 1.59 (1.90) | 2.09 (2.41) |
| Daily minimum temperature [°C] | 0.91 (1.50) | 1.09 (1.63) | 1.42 (1.85) | 1.81 (2.17) |
| Daily maximum wind speed [km/h] | 2.25 (3.60) | 2.70 (4.07) | 3.43 (4.76) | 4.39 (5.69) |
| Daily precipitation total [mm] | 4.15 (4.33) | 5.18 (5.24) | 6.80 (6.99) | 7.11 (7.30) |
Who wins where
A single global number hides a lot. The board breaks the record down by period, by lead time and by city. For every city it also lays out a rolling timeline: the last seven days with the actual value and the model that came closest, today so far, and the forecast for the next ten days. Each morning yesterday's forecast becomes today's result, and it is scored.
How Darwin-Prediction-Weather works
Darwin is VIDRAFT's family of models built on one idea: combine the strengths of several strong models into one that is stronger than any of them. Darwin-Prediction-Weather applies that idea to the atmosphere.
- Forecast engine. A single forecast from skill-weighted fusion of global forecasts, with local correction for temperature and wind. It learns only from days whose observed value was already known before the forecast it is scored on. Nothing from the future enters.
- Decision core. A Darwin-27B model with a calibrated readout turns forecast fields into decisions: go or no-go, threshold exceedance, category probabilities.
The same engine was evaluated on WeatherBench 2, the standard benchmark for AI weather models run by Google Research. On the 2020 deterministic protocol it records the lowest error in all 15 evaluated cells: 2 m temperature, 500 hPa geopotential and 850 hPa temperature at 1, 3, 5, 7 and 10 days. The forecast files are public and the entry has been submitted to the WeatherBench 2 team.
| WeatherBench 2, 2020 | 1 day | 3 days | 5 days | 7 days | 10 days |
|---|---|---|---|---|---|
| 2 m temperature [K] | 0.468 | 0.743 | 1.107 | 1.531 | 1.984 |
| best on official scorecard | 0.488 | 0.786 | 1.125 | 1.552 | 2.022 |
| Z500 [m2/s2] | 33.7 | 108.5 | 240.6 | 400.3 | 595.4 |
| best on official scorecard | 36.8 | 113.8 | 250.7 | 415.1 | 606.5 |
| T850 [K] | 0.503 | 0.901 | 1.406 | 2.000 | 2.699 |
| best on official scorecard | 0.518 | 0.922 | 1.435 | 2.041 | 2.737 |
Everything is open to check
- The board: https://huggingface.co/spaces/FINAL-Bench/Weather-Model-Leaderboard
- The daily record of every forecast and observed value since 2026-01-01, as CSV: https://huggingface.co/datasets/FINAL-Bench/Weather-Model-Forecast-Archive
- The model: https://huggingface.co/FINAL-Bench/Darwin-Prediction-Weather
- WeatherBench 2 forecast files: https://huggingface.co/datasets/VIDraft/Darwin-Prediction-Weather-WB2
- Full article: https://huggingface.co/blog/FINAL-Bench/which-weather-model-is-most-accurate-we-score-them
What comes next
Forecasts 8 to 10 days ahead enter the ranking as their target days are verified. Scoring against station observations will run alongside ERA5. More cities and more elements follow, and a forecasting game on Hugging Face will let anyone test their own forecast against the models.
License and sources
The leaderboard application is released under the Apache License 2.0. The compiled daily record and the Darwin-Prediction-Weather model are released under CC BY-NC 4.0, non-commercial use only. For commercial use, contact vidraft@mail.vidraft.net.
Weather data by Open-Meteo.com (CC BY 4.0). Verification: ERA5, Copernicus Climate Change Service. Model sources: ECMWF, NOAA NCEP, DWD, JMA, ECCC and MET Norway. Satellite imagery: NASA Global Imagery Browse Services (GIBS).
