Salesforce logo

Moirai-1.1-R-Large

online
Salesforce/moirai-1.1-R-large

~311M params | 512 context | $0.00025 per forecast

Moirai-1.1-R-Large is the highest-capacity checkpoint in the updated Moirai 1.1-R line, the top tier above the 1.1-R Small and Base variants. At roughly 311M parameters it matches the footprint of Moirai-1.0-R-Large while carrying the 1.1 update, giving the newer dense family its strongest member.

The official card does not publish a full new technical note, but the released configuration follows the same masked-encoder Moirai architecture and usage pattern as the 1.0-R design, so it supports multivariate forecasting, dynamic covariates, and calibrated quantile output. Salesforce frames 1.1-R as a direct improvement on the original 1.0-R family with better low-frequency behavior; the card does not restate a new corpus, so the official provenance is the 1.0-R/Moirai line plus the 1.1 release update.

On TSFM.ai pick it when you want the highest-capacity checkpoint in the updated dense family and accuracy outweighs serving cost, particularly for low-frequency yearly and quarterly series. Step down to Moirai-1.1-R-Base or Moirai-1.1-R-Small when throughput or cost is the constraint, or use the original 1.0-R large checkpoint when you specifically want the first-generation release.

Model Classification

Family

Moirai

Type

time series foundation model

Pretrained time-series model exposed on TSFM.ai for zero-shot or few-shot forecasting workloads.

Training Data

Salesforce positions this as an updated 1.0-R release; the 1.1 card does not restate a new corpus in detail, so the official provenance is the 1.0-R/Moirai line plus the 1.1 release update.

Recommended For

  • Multivariate forecasting across heterogeneous domains
  • Workloads that benefit from probabilistic outputs and arbitrary variate counts

Strengths

  • Strong multivariate coverage across the Moirai family
  • Well-suited to covariates and correlated series

Limitations

  • Model cards for some newer Moirai variants are still sparse on exact checkpoint details
  • Heavier family choices can be more expensive than tiny single-purpose baselines

Capabilities

forecastingquantile-forecastingmultivariatecovariateszero-shot

Tags

salesforcemoirailow-frequencyquality-tier

Specifications

Parameters
~311M
Architecture
updated masked encoder Moirai architecture following the 1.0-R design pattern
Context length
512
Max context
8,192
Minimum history
n/a
Recommended history
n/a
Input step
n/a
Required target series
1
Temperature
Ignored
Top P
Ignored
Max output
2,048
Avg latency
n/a
Uptime
n/a
Plan limits
1,000 rpm free · 1,000,000 rpm with billing
Accelerator
L40S
Regions
Virginia, US
License
n/a

Pricing

Per forecast
$0.00025

Performance

Average latency
n/a
Availability
n/a
Plan limits
1,000 rpm free · 1,000,000 rpm with billing

Related models