Models & methods

Zero-shot forecasting

Zero-shot forecasting produces a path for a series the model was not trained on, using a pretrained foundation model. You pass history at inference time. There is no local training job on that series first.

It is the cold-start button with a modern engine. Still score it. Zero-shot can lose to a well-tuned seasonal naive on a long, stable series — and that is fine; put both in the tournament.

Zero-shot is not 'the model never sees data.' It sees your history as input. It was not optimized on your history as training labels.