我们提出 TimesFM-3，一种前沿的时间序列基础模型，可在单次前向传播中实现高精度多元时间序列预测，在主要基准测试中显著优于其他预测模型。对于目标序列和历史协变量序列，每个 token 直接由单个 patch 构建；而对于历史-未来协变量序列，TimesFM-3 采用了一种巧妙的“前瞻”策略：每个 token 将当前 patch 与后续若干 future patch 拼接，使模型能够提前感知已知的未来信号。

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. 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.

我们推出 TimesFM-3，这是一款最先进的时序基础模型，可在单次前向传播中实现高精度的多元时间序列预测，在主要基准测试中显著优于其他预测模型。对于目标序列和历史协变量序列，每个 token 直接由单个 patch 构建；而对于历史-未来协变量，TimesFM-3 采用了一种巧妙的“前瞻”策略：每个 token 将当前 patch 与后续若干 future patch 拼接，使模型得以提前感知未来已知信号。上述两种注意力机制在若干网络层中交替使用，从而无缝融合时序模式与跨序列关系。促销规划：TimesFM-3 的多元预测利用促销协变量，以预估未来计划促销日的销售提升幅度。

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. 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. These two attention mechanisms alternate for several layers, seamlessly blending temporal patterns with cross-series relationships. Planning promotions: TimesFM-3's multivariate forecast uses a promotion covariate to anticipate sales lift on planned promotion days in the future.