Aller au contenu principal
Version: Suivant

OpenAIModerationsAPI

The OpenAIModerationsAPI is responsible for classifying if text and/or image inputs are potentially harmful.

https://platform.openai.com/docs/api-reference/moderations

Fonctions​

create()​

create(input : Variant; model : Text; parameters : OpenAIParameters) : OpenAIModerationResult

ParamètresTypeDescription
zone de saisieVariantInput (or inputs) to classify. Can be a single text or a collection of OpenAIMessage.
modelTextThe content moderation model you would like to use.
paramètresOpenAIParametersAdditional parameters for the request.
RésultatOpenAIModerationResultThe result of the moderation, indicating potential harm.

Classifies whether the input is potentially harmful.

https://platform.openai.com/docs/api-reference/moderations/create

Exemples​

Moderate a text​

var $result:=$client.moderation.create("Some text to classify"; "omni-moderation-latest"; $parameters)

Moderate a text and an image​

var $messages:=[{type: "text"; text: "...text to classify goes here..."}; \
                {type: "image_url"; image_url: {url: "https://example.com/image.png"}}]
var $result:=$client.moderation.create($messages; "omni-moderation-latest"; $parameters)