Skip to main content

Supported providers

OpenAI​

  • Base URL: https://api.openai.com/v1
  • Default model: omni-moderation-latest
  • API endpoint: POST /moderations

Models​

ModelDescription
omni-moderation-latestLatest multimodal moderation model
text-moderation-latestLatest text-only moderation model
text-moderation-stableStable text-only moderation model

Categories​

OpenAI moderation models classify content into the following categories:

CategoryDescription
hateContent expressing hate toward a group
hate/threateningHateful content with threats of violence
harassmentContent harassing individuals or groups
harassment/threateningHarassment with threats of violence
self-harmContent promoting or depicting self-harm
self-harm/intentContent expressing intent of self-harm
self-harm/instructionsInstructions for self-harm
sexualSexual content
sexual/minorsSexual content involving minors
violenceContent depicting violence
violence/graphicGraphic depictions of violence

Mistral​

  • Base URL: https://api.mistral.ai
  • Default model: mistral-moderation-latest
  • API endpoint: POST /v1/moderations

Models​

ModelDescription
mistral-moderation-latestLatest Mistral moderation model

Provider comparison​

ProviderDefault modelAPI formatCategories
OpenAIomni-moderation-latestOpenAI11 categories
Mistralmistral-moderation-latestOpenAI-compatibleProvider-specific