Documentation

Responses API

Explore this guide to learn how ChatboxAI works and how to integrate it into your workflow.

Responses API

The Responses API is a custom, higher-level interface exclusive to ChatboxAI. It simplifies multi-turn interactions, provides a cleaner request format, and introduces output_text for instant string access, saving you from parsing deep JSON structures.

1. Overview

While the standard chat/completions endpoint adheres strictly to the OpenAI specification, the Responses API is designed for developers who want a faster, less verbose integration.

Feature	Chat Completions	Responses API
Input Format	`messages[]` array of objects	`input` (Accepts a simple string OR array)
System Prompt	`messages[0].role = "system"`	Dedicated `instructions` field
Output Access	`choices[0].message.content`	`output_text` (Direct string access)
State Tracking	Client must store history	Server-side memory (Coming in v2)

2. Basic Request

Text

POST https://chatboxai.co.in/api/v1/responses

Instead of building a messages array, you can pass a single string to input and define the persona in instructions.

Example (Fetch API)

JAVASCRIPT

const response = await fetch("https://chatboxai.co.in/api/v1/responses", {
  method: "POST",
  headers: {
    "Authorization": `Bearer ${process.env.CHATBOXAI_API_KEY}`,
    "Content-Type": "application/json",
  },
  body: JSON.stringify({
    model: "chatboxai/llama-3.3-70b", // Free tier model
    instructions: "You are a helpful geography assistant. Be highly concise.",
    input: "What is the capital of France?",
  }),
});

const data = await response.json();

// Directly access the generated string!
console.log(data.output_text); 
// Output: "The capital of France is Paris."

3. The Response Object

The Responses API returns a simplified JSON object. The inclusion of cost_usd remains standard.

JSON

{
  "id": "resp_01abc123",
  "object": "response",
  "created": 1716123456,
  "model": "chatboxai/llama-3.3-70b",
  "output_text": "The capital of France is Paris.",
  "output": [
    {
      "type": "message",
      "role": "assistant",
      "content": "The capital of France is Paris."
    }
  ],
  "usage": {
    "input_tokens": 25,
    "output_tokens": 12,
    "total_tokens": 37
  },
  "cost_usd": 0.000000
}

4. Multi-turn Conversations

For multi-turn chat in v1, you still pass an array to input, but the format remains cleaner than standard completions.

JAVASCRIPT

const response = await fetch("https://chatboxai.co.in/api/v1/responses", {
  method: "POST",
  headers: { 
    "Authorization": `Bearer ${process.env.CHATBOXAI_API_KEY}`, 
    "Content-Type": "application/json" 
  },
  body: JSON.stringify({
    model: "chatboxai/claude-4.5-sonnet", // Paid model
    instructions: "You are a friendly AI.",
    input: [
      { role: "user", content: "My name is Alice." },
      { role: "assistant", content: "Nice to meet you, Alice!" },
      { role: "user", content: "What is my name?" },
    ],
  }),
});

const data = await response.json();
console.log(data.output_text); // "Your name is Alice."

5. Request Parameters Reference

Parameter	Type	Required	Description
`model`	string	Yes	The ID of the model (e.g., `chatboxai/qwen3-coder`).
`input`	string \| array	Yes	A single string (for 1-turn) or an array of message objects.
`instructions`	string	No	Sets the model's behavior/persona (System Prompt).
`tools`	array	No	Tool definitions for function calling.
`temperature`	number	No	Creativity control. Default `0.7`.
`max_output_tokens`	integer	No	Maximum generation length.