Toto: An Open Time Series Foundation Model Optimized for Observability

Ben Cohen, Emaad Khwaja, Youssef Doubli, Salahidine Lemaachi, Chris Lettieri, Charles Masson, Hugo Miccinilli, Elise Ramé, Qiqi Ren, Afshin Rostamizadeh, Jean Ogier du Terrail, Anna-Monica Toon, Kan Wang, Stephan Xie, Zongzhe Xu, Viktoriya Zhukova, David Asker, Ameet Talwalkar, Othmane Abou-Amal

Keywords: time series, foundation models, observability

TL;DR: We release a new open-weights time series foundation model (TSFM) with SOTA performance on all evaluated benchmarks.

Abstract: We introduce Toto, a time series forecasting foundation model with 151 million parameters. Toto uses a modern decoder-only architecture coupled with architectural innovations designed to account for specific challenges found in multivariate observability time series data. Toto's pre-training corpus is a mixture of observability data, open datasets, and synthetic data, and is 4-10$\times$ larger than those of leading time series foundation models. We source observability data exclusively from our own telemetry and internal observability metrics. Extensive evaluations demonstrate that \toto achieves state-of-the-art performance on general purpose time series forecasting benchmarks. Toto's model weights, inference code, and evaluation scripts are available as open source under the Apache 2.0 License.

Submission Number: 36