This Time is Different: An Observability Perspective on Time Series Foundation Models

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Cohen, Ben, Khwaja, Emaad, Doubli, Youssef, Lemaachi, Salahidine, Lettieri, Chris, Masson, Charles, Miccinilli, Hugo, Ramé, Elise, Ren, Qiqi, Rostamizadeh, Afshin, Terrail, Jean Ogier du, Toon, Anna-Monica, Wang, Kan, Xie, Stephan, Xu, Zongzhe, Zhukova, Viktoriya, Asker, David, Talwalkar, Ameet, Abou-Amal, Othmane
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908628823310336
author Cohen, Ben
Khwaja, Emaad
Doubli, Youssef
Lemaachi, Salahidine
Lettieri, Chris
Masson, Charles
Miccinilli, Hugo
Ramé, Elise
Ren, Qiqi
Rostamizadeh, Afshin
Terrail, Jean Ogier du
Toon, Anna-Monica
Wang, Kan
Xie, Stephan
Xu, Zongzhe
Zhukova, Viktoriya
Asker, David
Talwalkar, Ameet
Abou-Amal, Othmane
author_facet Cohen, Ben
Khwaja, Emaad
Doubli, Youssef
Lemaachi, Salahidine
Lettieri, Chris
Masson, Charles
Miccinilli, Hugo
Ramé, Elise
Ren, Qiqi
Rostamizadeh, Afshin
Terrail, Jean Ogier du
Toon, Anna-Monica
Wang, Kan
Xie, Stephan
Xu, Zongzhe
Zhukova, Viktoriya
Asker, David
Talwalkar, Ameet
Abou-Amal, Othmane
contents 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. Additionally, we introduce BOOM, a large-scale benchmark consisting of 350 million observations across 2,807 real-world time series. For both Toto and BOOM, we source observability data exclusively from Datadog's own telemetry and internal observability metrics. Extensive evaluations demonstrate that Toto achieves state-of-the-art performance on both BOOM and on established general purpose time series forecasting benchmarks. Toto's model weights, inference code, and evaluation scripts, as well as BOOM's data and evaluation code, are all available as open source under the Apache 2.0 License available at https://huggingface.co/Datadog/Toto-Open-Base-1.0 and https://github.com/DataDog/toto.
format Preprint
id arxiv_https___arxiv_org_abs_2505_14766
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle This Time is Different: An Observability Perspective on Time Series Foundation Models
Cohen, Ben
Khwaja, Emaad
Doubli, Youssef
Lemaachi, Salahidine
Lettieri, Chris
Masson, Charles
Miccinilli, Hugo
Ramé, Elise
Ren, Qiqi
Rostamizadeh, Afshin
Terrail, Jean Ogier du
Toon, Anna-Monica
Wang, Kan
Xie, Stephan
Xu, Zongzhe
Zhukova, Viktoriya
Asker, David
Talwalkar, Ameet
Abou-Amal, Othmane
Machine Learning
Artificial Intelligence
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. Additionally, we introduce BOOM, a large-scale benchmark consisting of 350 million observations across 2,807 real-world time series. For both Toto and BOOM, we source observability data exclusively from Datadog's own telemetry and internal observability metrics. Extensive evaluations demonstrate that Toto achieves state-of-the-art performance on both BOOM and on established general purpose time series forecasting benchmarks. Toto's model weights, inference code, and evaluation scripts, as well as BOOM's data and evaluation code, are all available as open source under the Apache 2.0 License available at https://huggingface.co/Datadog/Toto-Open-Base-1.0 and https://github.com/DataDog/toto.
title This Time is Different: An Observability Perspective on Time Series Foundation Models
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2505.14766