thuml/OpenLTM

Implementations, Pre-training Code and Datasets of Large Time-Series Models

45
/ 100
Emerging

This project provides an open environment for researchers and advanced data scientists to develop, evaluate, and fine-tune large time-series models (LTMs), also known as Time Series Foundation Models (TSFMs). You can input diverse time-series data, whether for pre-training or supervised tasks, and it outputs trained models capable of advanced forecasting or feature extraction. This is for users who want to push the boundaries of time-series analysis with state-of-the-art deep learning architectures.

526 stars. No commits in the last 6 months.

Use this if you are a researcher or advanced practitioner looking to experiment with, build, or adapt powerful deep learning models for complex time-series forecasting and analysis tasks.

Not ideal if you need a simple, out-of-the-box tool for routine time-series forecasting without deep model development or adaptation.

time-series-forecasting predictive-modeling deep-learning-research data-science-advanced foundation-models
Stale 6m No Package No Dependents
Maintenance 2 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 17 / 25

How are scores calculated?

Stars

526

Forks

51

Language

Jupyter Notebook

License

MIT

Last pushed

Jul 21, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/thuml/OpenLTM"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.