SAS Model Manager Features List

Unify model assets

Unify model assets

  • Establish a secure and versioned model registry. Enable your team to understand your projects, models, metadata and supporting artifacts.
  • Profile, tag, sort and categorize your modeling assets. Search and discover models across your organization.
  • Access models and artifacts using the user interface or programmatically through REST API, SAS code or Python code to integrate with automated MLOps processes.
  • Manage, version, score and govern SAS, Python, R and other models across your enterprise.
  • Audit major events including model creation, deletion and deployment.
  • Read the blog to learn more about versioning in SAS Model Manager.

Validate models

Validate models

  • Automatically generate scoring code for Python and R models using the sasctl package.
  • Ensure models run within SAS Model Manager and production environments.
  • Compare SAS, Python and R models side by side to determine the best fit for production.
  • Read the blogto learn how to register Python models using a new step in SAS Studio.

Deploy models in minutes

Deploy models in minutes

  • Deploy SAS, Python and R models across a variety of destinations, with no recoding.
  • Deploy models in the cloud, on premise, in Snowpark or in Azure Machine Learning to be included in databases like Teradata.
  • Publish SAS, Python and R models into containers saved on registries within the cloud or on premise.
  • Read our guide to deploying models using CAS.

Monitor, detect, alert and repeat

Monitor, detect, alert and repeat

  • Illuminate data, concept and model drift with ongoing monitoring.
  • Gain dynamic performance reporting with out-of-the-box reporting from deployment to retirement.
  • Track, validate and audit reports to select champion models for use in other applications.
  • Synchronize KPIs to alert stakeholders or cue model retraining, minimizing costly downtime.
  • Customize performance reports and create your own KPIs based on model performance data.
  • Read our guide to monitoring natural language processing models within SAS Model Manager.

Streamline MLOps processes

Streamline MLOps processes

  • Use templates to create repeatable Continuous Integration and Continuous Delivery (CI/CD) processes to efficiently test and promote new models and decision flows.
  • Integrate multiple environments, tools and applications through our flexible API ecosystem.
  • Automate, integrate, promote and keep the right people informed with an adaptable workflow ecosystem.
  • Read the blogto learn how to leverage a CI/CD Promotion template using SAS Model Manager.