# TiRex-2: Generalizing TiRex to Multivariate Data and Streaming

arXiv: https://arxiv.org/abs/2607.01204 · HTML: https://mlsanity.com/paper/2607.01204/

## Key facts

- Cited by 13 papers in the map (data through 14 Oct 2026).
- Builds on 38 of its 65 references that are in the map.
- First posted on arXiv 1 Jul 2026.

## Builds on (most cited first, 30 of 38)

- [A Time Series is Worth 64 Words: Long-term Forecasting with Transformers](https://mlsanity.com/paper/2211.14730/) (2022)
- [iTransformer: Inverted Transformers Are Effective for Time Series Forecasting](https://mlsanity.com/paper/2310.06625/) (2023)
- [Chronos: Learning the Language of Time Series](https://mlsanity.com/paper/2403.07815/) (2024)
- [A decoder-only foundation model for time-series forecasting](https://mlsanity.com/paper/2310.10688/) (2023)
- [Unified Training of Universal Time Series Forecasting Transformers](https://mlsanity.com/paper/2402.02592/) (2024)
- [Large Language Models Are Zero-Shot Time Series Forecasters](https://mlsanity.com/paper/2310.07820/) (2023)
- [xLSTM: Extended Long Short-Term Memory](https://mlsanity.com/paper/2405.04517/) (2024)
- [TimeXer: Empowering Transformers for Time Series Forecasting with Exogenous Variables](https://mlsanity.com/paper/2402.19072/) (2024)
- [TSMixer: An All-MLP Architecture for Time Series Forecasting](https://mlsanity.com/paper/2303.06053/) (2023)
- [GIFT-Eval: A Benchmark For General Time Series Forecasting Model Evaluation](https://mlsanity.com/paper/2410.10393/) (2024)
- [Sundial: A Family of Highly Capable Time Series Foundation Models](https://mlsanity.com/paper/2502.00816/) (2025)
- [PromptCast: A New Prompt-based Learning Paradigm for Time Series Forecasting](https://mlsanity.com/paper/2210.08964/) (2022)
- [Tiny Time Mixers (TTMs): Fast Pre-trained Models for Enhanced Zero/Few-Shot Forecasting of Multivariate Time Series](https://mlsanity.com/paper/2401.03955/) (2024)
- [TiRex: Zero-Shot Forecasting Across Long and Short Horizons with Enhanced In-Context Learning](https://mlsanity.com/paper/2505.23719/) (2025)
- [Moirai-MoE: Empowering Time Series Foundation Models with Sparse Mixture of Experts](https://mlsanity.com/paper/2410.10469/) (2024)
- [UniTS: A Unified Multi-Task Time Series Model](https://mlsanity.com/paper/2403.00131/) (2024)
- [Chaos as an interpretable benchmark for forecasting and data-driven modelling](https://mlsanity.com/paper/2110.05266/) (2021)
- [Timer-XL: Long-Context Transformers for Unified Time Series Forecasting](https://mlsanity.com/paper/2410.04803/) (2024)
- [Moirai 2.0: When Less Is More for Time Series Forecasting](https://mlsanity.com/paper/2511.11698/) (2025)
- [This Time is Different: An Observability Perspective on Time Series Foundation Models](https://mlsanity.com/paper/2505.14766/) (2025)
- [fev-bench: A Realistic Benchmark for Time Series Forecasting](https://mlsanity.com/paper/2509.26468/) (2025)
- [Toto: Time Series Optimized Transformer for Observability](https://mlsanity.com/paper/2407.07874/) (2024)
- [In-Context Fine-Tuning for Time-Series Foundation Models](https://mlsanity.com/paper/2410.24087/) (2024)
- [Timer-S1: A Billion-Scale Time Series Foundation Model with Serial Scaling](https://mlsanity.com/paper/2603.04791/) (2026)
- [Output Scaling: YingLong-Delayed Chain of Thought in a Large Pretrained Time Series Forecasting Model](https://mlsanity.com/paper/2506.11029/) (2025)
- [ChronosX: Adapting Pretrained Time Series Models with Exogenous Variables](https://mlsanity.com/paper/2503.12107/) (2025)
- [A Comprehensive Survey of Deep Transfer Learning for Anomaly Detection in Industrial Time Series: Methods, Applications, and Directions](https://mlsanity.com/paper/2307.05638/) (2023)
- [TempoPFN: Synthetic Pre-training of Linear RNNs for Zero-shot Time Series Forecasting](https://mlsanity.com/paper/2510.25502/) (2025)
- [AdaPTS: Adapting Univariate Foundation Models to Probabilistic Multivariate Time Series Forecasting](https://mlsanity.com/paper/2502.10235/) (2025)
- [Bio-xLSTM: Generative modeling, representation and in-context learning of biological and chemical sequences](https://mlsanity.com/paper/2411.04165/) (2024)

## Built on by (most cited first, 13 of 13)

- [Into the ORBIT for Time Series: Training Regimes for Foundation Models](https://mlsanity.com/paper/2608.13262/) (2026)
- [t0: A Time-Series Foundation Model for Forecasting with Context](https://mlsanity.com/paper/2609.24559/) (2026)
- [SANE: State Anomaly Neutralization for Stable Extreme-Context Delta-Rule Models](https://mlsanity.com/paper/2608.22354/) (2026)
- [Distillation of Synthetic Data for Time Series Foundation Models](https://mlsanity.com/paper/2609.09586/) (2026)
- [FlowTSFM: Turning Encoder Depth into Quantile Transport](https://mlsanity.com/paper/2609.13640/) (2026)
- [Aurora-X: Built for Extreme Time Series Forecasting](https://mlsanity.com/paper/2609.31038/) (2026)
- [Self-Evolving Time-Series Forecasting Agents with Episodic Memory and Online Policy Learning](https://mlsanity.com/paper/2609.32689/) (2026)
- [OpenTSLM TeeMoE: A Unified Time-Series Language Model for Forecasting, Contextual Prediction, and Reasoning](https://mlsanity.com/paper/2609.40265/) (2026)
- [Pooling Helps, Learned Weighting Hurts In-Context: Decomposing Group Attention](https://mlsanity.com/paper/2610.01831/) (2026)
- [A Unified Scaling Law for Time Series Foundation Models](https://mlsanity.com/paper/2610.05269/) (2026)
- [Timer-M1: A Multivariate Time Series Foundation Model via Learning Primitives](https://mlsanity.com/paper/2610.11734/) (2026)
- [On the Encoding of Calendar Features in Covariate-Informed Time Series Foundation Models](https://mlsanity.com/paper/2610.13676/) (2026)
- [TimeRSI: Towards Self-Improved Agentic Deployment for Time-Series Foundation Models](https://mlsanity.com/paper/2610.19725/) (2026)

## Abstract (arXiv)

We introduce TiRex-2, a recurrent xLSTM-based time series foundation model that generalizes the univariate TiRex to multivariate forecasting with both past and future covariates. Real-world forecasting is inherently sequential: observations arrive continuously, variables evolve jointly, and a subset of covariates is known ahead of time. Existing Transformer-based time series foundation models capture cross-variate dependencies but incur quadratic complexity in context length and require full-history recomputation as new observations arrive. TiRex-2 addresses these limitations through a memory-centric recurrent design that operates at constant per-patch cost under streaming. The model combines a bidirectional time mixer with an asymmetric grouped-attention variate mixer, enabling the integration of future-known covariates while preserving strict causality over target variables. To our knowledge, this is the first time series foundation model that achieves this combination of properties. To support scalable multivariate pretraining, we propose a synthetic coupling pipeline that composes diverse multivariate samples on the fly from large univariate corpora. Empirically, TiRex-2 achieves state-of-the-art zero-shot performance on GIFT-Eval and fev-bench, remains stable when streamed to arbitrary context lengths, and maintains constant inference cost per patch. The model uses 38.4M active parameters in univariate mode, with an additional 44.1M parameters activated for multivariate forecasting.