open-ai-api

Completion API

Endpoint

Endpoint URL: https://api.openai.com/v1/completions

Method: POST

Request Parameters

• engine (String): Required. ID of the engine for the task.
• prompt (String/Array): Optional. Text for completions generation.
• max_tokens (Integer): Optional. Maximum tokens in the completion. Default: 16.
• temperature (Float): Optional. Token sampling method. Default: 1.
• top_p (Float): Optional. Nucleus sampling. Default: 1.
• suffix (String): Optional. Text at the end for better generation.
• n (Integer): Optional. Number of completions. Default: 1.
• logprobs (Integer): Optional. Number of tokens with highest likelihood. Default: null.
• best_of (Integer): Optional. Server-side generates best_of completions. Default: 1.
• presence_penalty (Integer): Optional. Penalizes new tokens based on presence. Range: -2.0 to 2.0.
• frequency_penalty (Integer): Optional. Penalizes new tokens based on existing frequency. Range: -2.0 to 2.0.

Response

• choices (Array): Array of objects with logprobs and generated text.

Sample API Call

// Define endpoint URL
const endpointUrl = "https://api.openai.com/v1/completions";

// Define Header Parameters
const headerParameters = {
  "Authorization": "Bearer {{SECRET_KEY}}",
  "Content-Type": "application/json"
};

Link to Sample React Project README