English

Toto 2.0: Time Series Forecasting Enters the Scaling Era

Machine Learning 2026-05-20 v1 Artificial Intelligence

Abstract

We show that time series foundation models scale: a single training recipe produces reliable forecast-quality improvements from 4M to 2.5B parameters. We release Toto 2.0, a family of five open-weights forecasting models trained under this recipe. The Toto 2.0 family sets a new state of the art on three forecasting benchmarks: BOOM, our observability benchmark; GIFT-Eval, the standard general-purpose benchmark; and the recent contamination-resistant TIME benchmark. This report describes our experimental results and details the design decisions behind Toto 2.0: its architecture and training recipe, training data, and the u-muP hyperparameter transfer pipeline. All five base checkpoints are released under Apache 2.0.

Keywords

Cite

@article{arxiv.2605.20119,
  title  = {Toto 2.0: Time Series Forecasting Enters the Scaling Era},
  author = {Emaad Khwaja and Chris Lettieri and Gerald Woo and Eden Belouadah and Marc Cenac and Guillaume Jarry and Enguerrand Paquin and Xunyi Zhao and Viktoriya Zhukov and Othmane Abou-Amal and Chenghao Liu and Ameet Talwalkar and David Asker},
  journal= {arXiv preprint arXiv:2605.20119},
  year   = {2026}
}

Comments

Code: https://github.com/DataDog/toto Weights: https://huggingface.co/collections/Datadog/toto-20