Moderations
Learn how to use the moderations endpoint of OpenAI to tag speech with various flags like hate speech, violence, self-harm, etc.
The moderations endpoint
Moderations is an endpoint that OpenAI provides to flag text. This API endpoint analyzes the text provided and evaluates whether the content complies with OpenAI’s usage policies. After the API analyzes the text, it tells the user which flags fit the text and the confidence that the API has for each flag.
The following URL uses the POST method, which can be used to call the moderations endpoint.
https://api.openai.com/v1/moderations
Understanding the moderations endpoint
Let’s look at the moderations endpoint in more detail, reviewing the various models, the request parameters, and the response parameters.
Request parameters
Let’s look at the parameters that are required to make a request at the moderations endpoint.
Fields | Format | Type | Description |
| String | Optional | The ID of the engine to use for the request. |
| String/array | Required | The text that would be flagged in the pre-determined categories by the API. |