Skip to main content
Open Workspace Settings > Monitoring to configure the default schedule and selected AI models for the current workspace.
Monitoring configuration is unavailable on the Free plan. Existing data remains separate from the ability to schedule new runs.

​
Summary

The top of the page shows the number of active monitoring prompts, the number of selected models, and the Estimated credits per day. The estimate is a daily average based on each prompt’s schedule and the selected models. Credits are consumed when analyses run, so weekly prompts use their weekly credits on the scheduled day.

​
Automatic schedule

Choose the Default Frequency:
  • No monitoring;
  • Daily;
  • Weekly, with a day of the week;
  • Monthly, with a day of the month from 1 to 28, so that the date exists in every month.
This value applies to prompts set to Auto. Enable Auto-generate responses to answer new prompts as soon as they are created; otherwise, new prompts are collected at the next scheduled run. Select Save Schedule after changing these settings.

​
Choose AI models

Select the models available for monitoring in your workspace. Premium API models are enabled on request, with a tailored offer. Model changes are saved automatically.
  • Core models monitor public AI experiences as your audience sees them, and cost 1 credit per answer. The table also shows the insights collected for each one.
  • Premium models access additional models and versions through provider APIs, and are billed per model.
Badges explain availability: Beta, Legacy model (a previous-generation model still available at its original credits), Coming soon, Temporarily unavailable (your selection is saved and collection resumes automatically), and Not available in a country. A Chinese assistant has no country targeting. Model costs are listed in Providers and credits. Changing the model set affects future runs. It does not regenerate historical answers.

​
Launch an analysis

Use Launch analysis to run the monitoring prompts now. You can restrict the run to a topic, a tag, or a model. The run costs the credits of the generated answers.

​
Prompt-level overrides

Open Workspace Settings > Prompts to configure individual prompts. Each prompt can:
  • inherit the default with Auto;
  • use Daily, Weekly, Monthly, or No monitoring;
  • be paused or resumed;
  • use a boost, personas, its own models, or selected countries.

​
Boosts

A boost requests 2 to 20 answers per model and run for a prompt. The workspace also enforces the boosted-prompt limit shown in the interface. Boosted prompts consume credits for every generated answer. Use the estimate to review the combined effect of prompt frequencies, selected models, countries, personas, and boosts before saving.

​
Estimate limitations

The displayed credit estimate reflects the current configuration. Actual use depends on which scheduled or manual generations run and which answers complete.