Search⌘ K
Home
Courses
Using OpenAI API for Natural Language Processing in JavaScript

Completing Incomplete Content and Factual Responses

Explore how to use the OpenAI completions endpoint to effectively complete unfinished sentences and generate precise factual responses. Understand key parameters like temperature to control creativity and accuracy, and apply these techniques to practical scenarios such as creative writing and social media content.

We'll cover the following...

Overview

The chat completions endpoint can also be used to complete unattended (incomplete) sentences. For example, if we leave a sentence half-written, the chat completions endpoint can complete the sentence provided that it has the appropriate context. In this case, the value of temperature is usually kept low so that the completion does not disregard the previous context.

Factual responses
Factual responses

We input the incomplete content and request parameters into the OpenAI API’s model to complete that content.

Completions: complete incomplete content
Completions: complete incomplete content

Content completion

Here is an example of an incomplete sentence that we can provide to the API for content completion:

Although the tiger is a solitary beast and\n

Example

Let’s try the above example in the code widget below:

Javascript (babel-node)
// Define endpoint URL here
const endpointUrl = "https://api.openai.com/v1/chat/completions";
// Define Header Parameters here
const headerParameters = {
  "Authorization": "Bearer {{SECRET_KEY}}",
  "Content-Type": "application/json"
};
// Body Parameters
const bodyParameters = JSON.stringify({
  model: "gpt-3.5-turbo",
  messages: [{role:"system",content:"Complete the sentence."},
  {role:"user",content:"Although the tiger is a solitary beast and\n"}],
  temperature: 0.20,
  max_tokens: 64,
  top_p: 1
});
// Setting API call options
const options = {
  method: "POST",
  headers: headerParameters,
  body: bodyParameters
};
// Function to make API call
async function completeContent() {
  try {
    const response = await fetch(`${endpointUrl}`, options);
    // Printing response
    printResponse(response);
  } catch (error) {
    // Printing error message
    printError(error);
  }
}
// Calling function to make API call
completeContent();

If we ...