TimesFM-3: A zero-shot foundation model for multivariate forecasting

We introduce TimesFM-3, a state-of-the-art time series foundation model that enables highly accurate multivariate time series forecasting in a single forward pass, significantly outperforming other forecasting models across major benchmarks.

Since the debut of TimesFM in 2024, we’ve seen the adoption of time-series foundation models for real-world time-series forecasting tasks across multiple domains, such as retail, finance, observability, manufacturing, healthcare and natural sciences.

Up until TimesFM-2.5 (released in September 2025), our models were strictly limited to univariate forecasting: forecasting using only the history of a single time series. Yet, most real-world forecasting problems are inherently multivariate: where multiple time series and auxiliary external features jointly impact the future forecast of a time series. Consider forecasting ice cream sales for a retail chain. Past sales alone rarely tell the full story. A good forecast should also draw on sales of related products (e.g., ice cream cones, syrups), historical foot traffic, and known future events like weather forecasts, promotions, and holidays.

Today we introduce TimesFM-3, the next generation of our time-series foundation model that is natively pre-trained for multivariate forecasting. TimesFM-3 has 330 million parameters and is pre-trained on a real-world and synthetic time-series corpus comprising more than 1 trillion time points. Building on the efficiency and zero-shot generalization of its predecessors, TimesFM-3 adds robust support for complex multivariate scenarios in a zero-shot manner. It can jointly predict multiple coevolving time series, capturing dependencies that improve overall accuracy without requiring task-specific fine-tuning. The model natively supports:

• Multiple targets: Forecast multiple related time series simultaneously (e.g., jointly forecasting different brands of ice cream). The model supports both point and quantile forecasts for all targets.
• Past covariates: Incorporate features that are only known historically (e.g., past foot traffic).
• Past-future (dynamic) covariates: Leverage known future events to guide the forecast (e.g., planned promotional campaigns or weather forecasts).

Under the hood: Architecture & inference

TimesFM-3 builds on the proven decoder-only transformer architecture of its predecessors. As in previous versions, we process time series efficiently by grouping contiguous data points into patches of 32 time steps. We then apply normalization per time-series similar to that of TimesFM-2.5 in order to account for time series with vastly different scales.

Multivariate token construction

For target and past-covariate series, a token is constructed directly from a single patch. However, for past-future covariates, TimesFM-3 employs a clever "lookahead" strategy: each token concatenates the current patch with future patches, allowing the model to peek at upcoming known signals.

Alternating attention architecture

Once the patches are tokenized, they pass through an input residual block and enter the main transformer stack, which operates as a 2D grid:

1. Causal temporal attention: Tokens attend horizontally across time. To prevent data leakage, this attention is strictly causal — a token can only look at past tokens within its own specific time series.
2. Full variate attention: Tokens attend vertically across series. At any given time step, a token can look at all other time series in the dataset, allowing the model to learn complex cross-series correlations (e.g., how a promotion in one series affects sales in another).

These two attention mechanisms alternate for several layers, seamlessly blending temporal patterns with cross-series relationships.

Non-autoregressive decode: Forecasting in a single pass

Previous versions of TimesFM generated forecasts one patch at a time, introducing latency, compounding error accumulation, and computational cost. TimesFM-3 uses the strategy of Contiguous Patch Masking to generate the entire forecasting horizon in a single forward pass. The model appends masked placeholder tokens for the future horizon alongside the observed context. Target and past-covariate series are masked in the horizon (since their future values are unknown), while past–future covariates remain visible, providing the model with known future signals like holidays or scheduled events. Through the alternating attention layers, the model fills in all masked horizon patches simultaneously, with no iterative loop required. The model predicts 9 quantiles (from the 10th to the 90th percentile) for each target time series at every horizon step, providing a full probabilistic view of the forecast uncertainty.

Illustrative example for multivariate forecasting

Let’s revisit the ice cream sales example. Imagine you are working on next month’s promotion schedule and want to forecast the sales to anticipate. A standard univariate model (the red line, below) looks at the historical sales and projects a weekly pattern forward — but it has no idea about planned promotions on specific days. TimesFM-3's multivariate mode (the blue line, below) takes a different approach: by passing in the planned promotion schedule as a past-future covariate, the model learns the relationship between promotions and sales lift from the historical context, then applies that knowledge to future days with planned promotions. The result is a forecast that anticipates a ~20% sales bump on each promotion day. In the chart below, the amber blocks in the promotion covariates highlight which days have promotions — and the blue forecast visibly responds to each one, while the red forecast does not. Over the full month, this adds up to a more accurate forecast for projected revenue.

Evaluation and benchmarks

We evaluated TimesFM-3 on three comprehensive public forecasting benchmarks: Gift-Eval, FEV-Bench, and Time. On all three benchmarks, TimesFM-3 is the top-ranked model in terms of both point and probabilistic forecasting metrics among all pre-trained foundation models. The plots below show average rank across tasks for both point forecast accuracy and probabilistic forecast quality (lower is better) for the three benchmarks. We compare against recent foundation models including multivariate-capable models, such as Chronos-2 and the Toto 2.0 family, as well as our previous model TimesFM-2.5.

Each plot includes two entries for TimesFM-3. The "univariate mode" point shows performance when the model is evaluated without any covariate or cross-series information, treating each target series independently, just like a traditional univariate model. Even in this univariate mode, TimesFM-3 already matches or outperforms other competing models. When we switch to the full multivariate mode, TimesFM-3 takes another leap, achieving the best average rank in both point and probabilistic forecasting across the board.

Conclusion

We introduce TimesFM-3, the latest generation of our TimesFM family of zero-shot time series foundation models, that obtains state-of-the-art multivariate and univariate forecasting performance on multiple public benchmarks. TimesFM-3 is now available on GitHub and Hugging Face, and is directly integrated into BigQuery. You can try TimesFM 3.0 immediately on both your multivariate and univariate tasks using theAI.FORECAST command in BigQuery — no ML expertise required.

Acknowledgements

This project is joint work with Yichen Zhou, Petros Mol, Abhimanyu Das and Samet Oymak.