Moirai-1.1-R-Base

Moirai-1.1-R-Base is the mid-size checkpoint in the updated Moirai 1.1-R family, sitting between the 1.1-R Small and Large variants. At roughly 91M parameters it matches the footprint of Moirai-1.0-R-Base while carrying the 1.1 update, making it the natural balanced default for the newer dense line.

It keeps the familiar masked-encoder Moirai architecture and usage pattern inherited from the 1.0-R design, so it handles multivariate forecasting, dynamic covariates, and calibrated quantile output the same way as the original family. Salesforce positions 1.1-R as an updated 1.0-R release rather than a new corpus: the card does not restate a fresh training mixture, so the official provenance is the 1.0-R/Moirai line plus the 1.1 release update, with the note emphasizing improved low-frequency performance.

On TSFM.ai choose it when you want Moirai's dense design but prefer the newer 1.1 checkpoints over the original 1.0 release, and want more accuracy than Moirai-1.1-R-Small without the cost of Moirai-1.1-R-Large. Step down to the small variant when throughput dominates, or up to the large variant for the highest-capacity checkpoint in the updated family.