Text → Moderate
AI-generatedSummary
Classify text content for policy violations using Unitalk AI's moderation models.
Inputs
- input (required) — The text to classify for policy violations.
- model (required) — The moderation model to use: 'text-moderation-latest' (default) or 'text-moderation-stable'.
- simplifyOutput — If true (default), wraps the raw response results in a 'data' property for easier access.
- requestOptions — Advanced HTTP options including batching (batch size and interval), SSL settings, proxy, and timeout.
Output shape
One item per input item, each containing a 'data' property with the classification results.
When 'simplifyOutput' is true (default), the output wraps the raw API results in a 'data' field. If disabled, the full API response body is returned.
Examples
Example 1: Moderate a list of user comments in batches of 50 with a 1-second delay between batches.
Input: List of comments | Model: text-moderation-latest | Simplify: true | RequestOptions: Batching enabled with batchSize 50 and batchInterval 1000
Example 2: Check a single message for policy violations and return the full raw API response.
Input: Single message string | Model: text-moderation-latest | Simplify: false