Moirai-1.1-R-Small

Moirai-1.1-R-Small is the smallest checkpoint in the updated Moirai 1.1-R line, the lightweight tier alongside the 1.1-R Base and Large variants. At roughly 14M parameters it mirrors the footprint of Moirai-1.0-R-Small while carrying the 1.1 release update, making it the cheapest entry point into the newer dense family.

The public card is sparse, but the released configuration follows the same overall masked-encoder Moirai architecture and API pattern as the 1.0-R family, so it forecasts multivariate series, accepts dynamic covariates, and returns calibrated quantiles in the same way. Salesforce positions 1.1-R as an updated 1.0-R release rather than a corpus change: the card does not restate a new training mixture, so the official provenance is the 1.0-R/Moirai line plus the 1.1 update.

On TSFM.ai reach for it when low-frequency yearly and quarterly series matter and you want the lowest-cost member of the updated family, since Salesforce frames 1.1-R as improving on 1.0-R specifically on those cases. Step up to Moirai-1.1-R-Base or Moirai-1.1-R-Large for more capacity within the same updated line, or fall back to the original 1.0-R checkpoints when you specifically want the first-generation release.