GIFT-Eval

GIFT-Eval evaluates general time-series forecasting models across diverse datasets, frequencies, domains, forecast horizons, and univariate/multivariate settings.

21rows
average_rankprimary metric
2026-05-06sampled

Metadata

ID
gift_eval
Category
Data
Release
2024-10-14

Metrics

Average Rank (lower is better), Average MAPE (lower is better), Average CRPS (lower is better)

Latest Results

Rows are ranked by the mean of source univariate and multivariate rank values from grouped_results_by_univariate.csv. Domain, frequency, and term-length grouped metrics are retained in metadata.

Rank Subject Average Rank Model Match Provenance Sampled
1 patch_tst 5.65 — Imported 2026-05-06
2 moirai_1.1_R_large_no_leak 6.06 — Imported 2026-05-06
3 i_transformer 6.25 — Imported 2026-05-06
4 tft 6.78 — Imported 2026-05-06
5 moirai_1.1_R_base_no_leak 7.44 — Imported 2026-05-06
6 moirai_1.1_R_small_no_leak 7.66 — Imported 2026-05-06
7 chronos_base 8.43 — Imported 2026-05-06
8 chronos_large 8.44 — Imported 2026-05-06
9 timesfm 8.70 — Imported 2026-05-06
10 chronos-small 9.20 — Imported 2026-05-06
11 tide 11.55 — Imported 2026-05-06
12 deepar 12 — Imported 2026-05-06
13 visionts 12.63 — Imported 2026-05-06
14 n_beats 13.19 — Imported 2026-05-06
15 auto_arima 13.23 — Imported 2026-05-06
16 crossformer 13.55 — Imported 2026-05-06
17 d_linear 14.36 — Imported 2026-05-06
18 auto_theta 15.18 — Imported 2026-05-06
19 auto_ets 15.95 — Imported 2026-05-06
20 seasonal_naive 16.63 — Imported 2026-05-06
21 naive 18.14 — Imported 2026-05-06