Create

API Reference

Moderations

Create

moderations.create(ModerationCreateParams**kwargs) -> modelstrresultslistModerationCreateResponse

post/moderations

Classifies if given messages are potentially harmful across several categories.

Parameters

messagesiterable

Iterable[MessageParam]

List of messages in the conversation.

Hide ParametersShow Parameters

contentunionroleliteralUserMessage

contentunionroleliteralSystemMessage

contentunionroleliteraltool_call_idstrToolResponseMessage

roleliteralcontentContentstop_reasonliteraltool_callslistCompletionMessage

modelstr

optional

Optional identifier of the model to use. Defaults to "Llama-Guard".

Returns

ModerationCreateResponseclass

Hide ParametersShow Parameters

modelstr

resultslist

List[Result]

Hide ParametersShow Parameters

flaggedbool

flagged_categorieslist

List[str]

HTTPTypeScriptPythonGo

from llama_api_client import LlamaAPIClient

client = LlamaAPIClient(
    api_key="My API Key",
)
moderation = client.moderations.create(
    messages=[{
        "content": "string",
        "role": "user",
    }],
)
print(moderation.model)

200 Example

{
  "model": "model",
  "results": [
    {
      "flagged": true,
      "flagged_categories": [
        "string"
      ]
    }
  ]
}