Skip to main content

​
Type Definition

type AnswerRequest = {
  message: string;
  model?: "gpt-4" | "gpt-4-turbo" | "gpt-3.5-turbo";
  maxTokens?: number;
  temperature?: number;
  instructions?: string;
  conversationId?: string;
  previousMessages?: Message[];
};

​
Properties

​
message
string
required
The user’s message or question to process. This is the main input that will be enriched with brand information.
​
model
string
The OpenAI model to use for generation. Available options:
  • "gpt-4" - Most capable model
  • "gpt-4-turbo" - Faster GPT-4 variant
  • "gpt-3.5-turbo" - Fastest, most cost-effective
Default: Uses the default model configured in your Thred client.
​
maxTokens
number
Maximum number of tokens to generate in the response. This value may be overridden by brand-specific settings.Default: Determined by brand settings or API defaults.
​
temperature
number
Sampling temperature for response generation (0-2). Lower values make output more focused and deterministic, higher values make it more creative and random. This value may be overridden by brand-specific settings.Default: Determined by brand settings or API defaults.
​
instructions
string
Additional instructions to guide the AI response. Use this to customize tone, format, or specific requirements for the generated response.
​
conversationId
string
Unique identifier to track conversation context across multiple requests. Use the same ID for follow-up messages in a conversation.
​
previousMessages
Message[]
Array of previous messages in the conversation for context. Each message should have a role (“user” or “assistant”) and content (string).

​
Examples

​
Basic Request

const request: AnswerRequest = {
  message: "What are the best running shoes for beginners?"
};

const response = await thred.answer(request);

​
Request with Model Selection

const request: AnswerRequest = {
  message: "Compare different types of protein powder",
  model: "gpt-4-turbo",
  temperature: 0.7,
  maxTokens: 500
};

const response = await thred.answer(request);

​
Request with Conversation Context

const request: AnswerRequest = {
  message: "What about for wide feet?",
  conversationId: "conv-123",
  previousMessages: [
    {
      role: "user",
      content: "What are the best running shoes for beginners?"
    },
    {
      role: "assistant",
      content: "For beginners, I recommend Nike Pegasus or Brooks Ghost..."
    }
  ]
};

const response = await thred.answer(request);

​
Request with Custom Instructions

const request: AnswerRequest = {
  message: "Tell me about electric bikes",
  instructions: "Keep the response under 100 words and focus on commuting benefits",
  model: "gpt-3.5-turbo"
};

const response = await thred.answer(request);

Build docs developers (and LLMs) love

Get started for freeTalk to us