Skip to main content
POST
Use omni-moderation-latest to perform safety moderation on input content. This model belongs to the Moderation Series and is not part of the image, video, or audio generation series.

Supported Models

ModelDescriptionSupported Inputs
omni-moderation-latestGeneral-purpose content moderation modelText, image, mixed text+image

Authorizations

Authorization
string
required
All endpoints require authentication using a Bearer Token.Get an API Key:Visit the API Key management page to obtain your API Key.Add the following header in your request:

Body

model
string
required
Moderation model name.
  • omni-moderation-latest - General-purpose content moderation model
input
string | string[] | object[]
required
Content to be moderated. Supports plain text, text array, or content block array.Content blocks can include:
  • text - Text content block
  • image_url - Image URL content block
stream
boolean
default:"false"
Whether to stream the response.
  • false: Non-streaming response (default; the only value currently supported, true is not supported)

input Request Modes

Mixed Text + Image

Plain Text (Single)

Plain Text (Array)

Image Only (Image URL)

Image Only (base64 Data URI)

Notes

  1. When input is a content block array, each element uses type to distinguish content type.
  2. For image moderation, prefer using publicly accessible URLs. If using base64, follow the standard Data URI format: data:image/{format};base64,{data}.
  3. Unless required otherwise, set stream: false consistently.