Create

API Reference

Moderations

Create

client.moderations.create(ModerationCreateParamsbody, RequestOptionsoptions?): modelstringresultsarrayModerationCreateResponse

post/moderations

Classifies if given messages are potentially harmful across several categories.

Parameters

bodymessagesarraymodelstringModerationCreateParams

Hide ParametersShow Parameters

messagesarray

Array<Message>

List of messages in the conversation.

Hide ParametersShow Parameters

contentunionrole"user"UserMessage

contentunionrole"system"SystemMessage

contentunionrole"tool"tool_call_idstringToolResponseMessage

role"assistant"contentunionstop_reasonuniontool_callsarrayCompletionMessage

modelstring

optional

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

Returns

ModerationCreateResponse

Hide ParametersShow Parameters

modelstring

resultsarray

Array<Result>

Hide ParametersShow Parameters

flaggedboolean

flagged_categoriesarray

Array<string>

HTTPTypeScriptPythonGo

import LlamaAPIClient from 'llama-api-client';

const client = new LlamaAPIClient({
  apiKey: 'My API Key',
});

const moderation = await client.moderations.create({ messages: [{ content: 'string', role: 'user' }] });

console.log(moderation.model);

200 Example

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