I like how the monitoring layer for bias, drift, and LLM-specific metrics in IBM watsonx.governance fires in real-time once a model is connected, with alerts that go straight to Slack instead of waiting on the next batch job. The factsheet/audit trail system is another standout, turning approval history and model documentation into exportable and usable formats for auditors, saving the team from having to rebuild it by hand. I find the combination of live monitoring and audit-ready output incredibly time-saving. The 'use case' structure grouping models under a shared business problem is more useful than expected, providing a common reference point for risk and data science teams. Integration with watsonx.ai is seamless, and the Lite plan allowing exploration before commitment is a big plus. The alerting setup efficiently routes issues to channels the team already monitors, ensuring timely triage. Review collected by and hosted on G2.com.

I find integrating models outside the IBM ecosystem, like through AWS SageMaker, requires manual configuration and the documentation isn't very clear. I notice the costs rise significantly once you scale into production. Also, the UI can feel a bit heavy, with slow loading when the inventory holds a lot of use cases. Review collected by and hosted on G2.com.