TimeCopilot
Azul Garza, Reneé Rosillo · 2025
Read TimeCopilot by Azul Garza, Reneé Rosillo (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2509-00616.txt.
Description
Azul Garza and Renee Rosillo introduce TimeCopilot. TimeCopilot is an open-source framework for time series forecasting. Forecasting means predicting future values from past patterns. The framework combines several Time Series Foundation Models with large language models. Foundation models here are large pre-trained forecasters. All parts connect through a single unified interface.
TimeCopilot automates the full forecasting pipeline for its users. It handles feature analysis, model choice, and cross-validation. Cross-validation checks a model on held-out slices of data. It then generates the forecast itself. Users also get explanations in plain language. They can ask direct questions about the future as well. On the GIFT-Eval benchmark, the system reaches top scores. It does so at low cost, the team reports.
Catalog record
| CREATOR | Azul Garza, Reneé Rosillo |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 13 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2509-00616