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

MOMENT: A Family of Open Time-series Foundation Models, ICML'24

A Gym for Agentic LLMs

[NeurIPS2025] MM-Agent: LLM as Agents for Real-world Mathematical Modeling Problem

A Library for Advanced Deep Time Series Models for General Time Series Analysis.

Official PyTorch implement of CIKM '25 paper: BALM-TSF

Official PyTorch repository for GRAM

PyTorch implementation of "ChatTime: A Unified Multimodal Time Series Foundation Model Bridging Numerical and Textual Data" (AAAI 2025 [oral])

Code for our paper "VisionTS: Visual Masked Autoencoders Are Free-Lunch Zero-Shot Time Series Forecasters".

[VLDB' 25] ChatTS: Understanding, Chat, Reasoning about Time Series with TS-MLLM

verl: Volcano Engine Reinforcement Learning for LLMs

An open-source implementaion for fine-tuning Qwen-VL series by Alibaba Cloud.

This repository provides the official implementation of ITFormer, a novel framework for temporal-textual multimodal question answering (QA), as presented in our ICML 2025 paper.

Official code for paper "TimeMaster: Training Time-Series Multimodal LLMs to Reason via Reinforcement Learning"

Source code for the AAAI 2025 paper "TimeCAP: Learning to Contextualize, Augment, and Predict Time Series Events with Large Language Model Agents."

Retrieval-Augmented Forecasting of Time-series

Implementation of the paper "In-context Time Series Predictor" (ICLR 2025)

Time-R1 is a two-stage reinforcement fine-tuning framework that trains large language models to perform slow-thinking, step-by-step reasoning for accurate and explainable time series forecasting.

[NeurIPS'24] Identifying Spatio-Temporal Drivers of Extreme Events

(ICLR 2022 Spotlight) Official PyTorch implementation of "How Do Vision Transformers Work?"

The repository documents code for "Validating Deep-Learning Weather Forecast Models on Recent High-Impact Extreme Events " https://doi.org/10.1175/AIES-D-24-0033.1

Source code for ClimateLearn