ChaosBench

Subseasonal-to-seasonal climate prediction benchmark with deterministic and probabilistic model leaderboards across T-850, Z-500, and Q-700 variables.

12rows
rmse_t_850_day44primary metric
2026-05-28sampled

Metadata

Metrics

T-850 RMSE at day 44 (lower is better), T-850 RMSE at day 28 (lower is better), T-850 RMSE at day 14 (lower is better), Z-500 RMSE at day 44 (lower is better), Q-700 RMSE at day 44 (lower is better), T-850 ACC at day 44, Z-500 ACC at day 44, Q-700 ACC at day 44, T-850 CRPSS at day 44, Z-500 CRPSS at day 44, Q-700 CRPSS at day 44

Latest Results

Rows are imported from the official ChaosBench Plotly iframe HTML files. Each result is a model/mode row, ranked by day-44 T-850 RMSE where lower is better.

Rank Subject T-850 RMSE at day 44 Model Match Provenance Sampled
1 climatology vs. ERA5 (control) 3.3882 T-850 RMSE at day 44 Imported 2026-05-28
2 ecmwf vs. ERA5 (ensemble) 3.4618 T-850 RMSE at day 44 Imported 2026-05-28
3 ncep vs. ERA5 (ensemble) 3.7616 T-850 RMSE at day 44 Imported 2026-05-28
4 ukmo vs. ERA5 (ensemble) 4.0983 T-850 RMSE at day 44 Imported 2026-05-28
5 cma vs. ERA5 (ensemble) 4.2579 T-850 RMSE at day 44 Imported 2026-05-28
6 fourcastnetv2 vs. ERA5 (control) 4.6793 T-850 RMSE at day 44 Imported 2026-05-28
7 ecmwf vs. ERA5 (control) 4.7230 T-850 RMSE at day 44 Imported 2026-05-28
8 ncep vs. ERA5 (control) 4.9005 T-850 RMSE at day 44 Imported 2026-05-28
9 ukmo vs. ERA5 (control) 4.9957 T-850 RMSE at day 44 Imported 2026-05-28
10 panguweather vs. ERA5 (control) 5.0215 T-850 RMSE at day 44 Imported 2026-05-28
11 graphcast vs. ERA5 (control) 5.0782 T-850 RMSE at day 44 Imported 2026-05-28
12 cma vs. ERA5 (control) 5.0815 T-850 RMSE at day 44 Imported 2026-05-28