Sundial (Base 128M)

Sundial is THUML's native generative TSFM for continuous-valued forecasting. The official card presents it as a 128M-parameter model built to generate multiple plausible futures rather than a single trajectory, making it one of the most directly probabilistic models in the hosted catalog and a strong option when distribution quality matters as much as point accuracy.

Architecturally it is a causal decoder-only transformer trained with a TimeFlow / flow-matching loss, which lets it model continuous values natively instead of relying on the discrete tokenization used by many other forecasting families. It is pretrained on TimeBench, a corpus of 1032B time points, with the model card listing THUML's UTSD, LOTSA, and the Chronos datasets as core upstream sources.

On TSFM.ai reach for Sundial when you want native generative probabilistic forecasting and care about the quality of the predictive distribution, not just a point estimate. Its generative focus can be overkill if you only need a single deterministic forecast, in which case a lighter point-forecast-first model is a better fit.