TEMPO

TEMPO is a GPT-based time-series foundation model from USC that takes an unusual route to zero-shot forecasting: instead of learning a bespoke time-series backbone, it adapts a pre-trained language model to the task. The result is an 80M-parameter forecaster that leans on the pattern-matching strength a GPT-2 backbone already carries.

Architecturally it pairs that GPT-2 backbone with trend-seasonal decomposition prompting. Input series are decomposed into trend, seasonal, and residual components, which lets the backbone reason about each structural piece separately before recombining them into a forecast. The official release is centered on the TEMPO-80M checkpoint and ships benchmark assets for ETT, electricity, traffic, and weather experiments on top of the GPT-2 base model, and it is released under Apache-2.0.

On TSFM.ai reach for TEMPO when your series have clear trend-seasonal structure that the decomposition can exploit, and when you want zero-shot forecasting without fine-tuning on downstream time-series data. The approach assumes meaningful trend-seasonal structure is present, and the GPT-2 backbone is larger than purpose-built time-series architectures at equivalent accuracy, so it is less compelling on series that lack that structure or when a leaner specialist forecaster would do.