TimeGPT forecasts all 145,063 Wikipedia web traffic series in 6.6 minutes, beating a top Kaggle statistical baseline by 18.8% on SMAPE, with zero feature engineering.
The page presents guides, tutorials, and insights on time series forecasting, anomaly detection, and predictive analytics, drawing on Nixtla's research and applications of TimeGPT.
This release introduces three major capabilities that together expand Nixtla from a single-model offering into a full time series intelligence platform
The page compares TimeGPT-2.1 with Databricks AutoML on electricity, weather, and retail demand datasets, measuring accuracy, runtime, and per-series win rates. TimeGPT-2.1 improved MAE, RMSE, MAPE, and WAPE across all three benchmarks.
Nixtla provides open-source tools and the TimeGPT platform using machine learning and statistical models to help businesses reduce uncertainty and make data-driven decisions.
