Package Exports
This package does not declare an exports field, so the exports above have been automatically detected and optimized by JSPM instead. If any package subpath is missing, it is recommended to post an issue to the original package (chatgpt-optimized-official) to support the "exports" field. If that is not possible, create a JSPM override to customize the exports field for this package.
Readme
chatgpt-optimized-official - a simple library to create ChatGPT that uses OpenAI official API
A simple Node.js module for creating ChatGPT using OpenAI official API.
Installation
To install the package, run the following command:
npm install chatgpt-optimized-officialOfficial Chat Completion API from OpenAI
import { ChatGPT } from "chatgpt-optimized-official";
let bot = new ChatGPT("<OPENAI_API_KEY>");
let response = await bot.ask("Hello?");
console.log(response);Creating ChatGPT with Official OpenAI Completion API
import { OpenAI } from "chatgpt-optimized-official";
let bot = new OpenAI("<OPENAI_API_KEY>");
let response = await bot.ask("Hello?");
console.log(response);Usage: Official Chat Completion API from OpenAI
import { ChatGPT } from "chatgpt-optimized-official";
let options = {
temperature: 0.7, // OpenAI parameter
max_tokens: 100, // OpenAI parameter [Max response size by tokens]
top_p: 0.9, // OpenAI parameter
frequency_penalty: 0, // OpenAI parameter
presence_penalty: 0, // OpenAI parameter
instructions: `You are ChatGPT, a large language model trained by OpenAI.`, // initial instructions for the bot
model: "gpt-3.5-turbo", // OpenAI parameter `gpt-3.5-turbo` is PAID
};
let bot = new ChatGPT("<OPENAI_API_KEY>", options); // Note: options is optional
let response = await bot.ask("Hello?");
console.log(response);
let conversationId = "conversation name";
let response1 = await bot.ask("Hello?", conversationId);
console.log(response1);
let conversationId2 = "another conversation name";
let response2 = await bot.ask("Hello?", conversationId2);
console.log(response2);Usage: Official Chat Completion API from OpenAI with functions
import { ChatGPT } from "chatgpt-optimized-official";
let options = {
temperature: 0.7, // OpenAI parameter
max_tokens: 100, // OpenAI parameter [Max response size by tokens]
top_p: 0.9, // OpenAI parameter
frequency_penalty: 0, // OpenAI parameter
presence_penalty: 0, // OpenAI parameter
instructions: `You are ChatGPT, a large language model trained by OpenAI.`, // initial instructions for the bot
model: "gpt-3.5-turbo-0613", // OpenAI parameter `gpt-3.5-turbo` is PAID
functions: [
{
"name": "saveDataUser",
"description": "Save data",
"parameters": {
"type": "object",
"properties": {
"name": {
"type": "string",
"description": "name user",
},
"email": {
"type": "string",
"description": "email user",
},
"phone": {
"type": "string",
"description": "phone user",
}
},
"required": ["name", "email", "phone"],
},
}
],
function_call: "auto",
};
let bot = new ChatGPT("<OPENAI_API_KEY>", options); // Note: options is optional
let response = await bot.askV1("Hello?");
console.log(response);
let conversationId = "conversation name";
let type = 1 // type 1 is user, 2 is assistant this is optional 1 (User) is default
let userName="User" //default value this is optional, username is other id for the conversation
let response1 = await bot.askV1("Hello?", conversationId, type, userName);
console.log(response1);
Usage: Creating ChatGPT with Official OpenAI Completion API
import { OpenAI } from "chatgpt-optimized-official";
let options = {
temperature: 0.7, // OpenAI parameter
max_tokens: 256, // OpenAI parameter [Max response size by tokens]
top_p: 0.9, // OpenAI parameter
frequency_penalty: 0, // OpenAI parameter
presence_penalty: 0, // OpenAI parameter
instructions: `You are ChatGPT, a large language model trained by OpenAI.`, // initial instructions for the bot
model: "text-davinci-003", // OpenAI parameter `text-davinci-003` is PAID
stop: "<|im_end|>", // OpenAI parameter
};
let bot = new OpenAI("<OPENAI_API_KEY>", options); // Note: options is optional
let response = await bot.ask("Hello?");
console.log(response);
let conversationId = "conversation name";
let response1 = await bot.ask("Hello?", conversationId);
console.log(response1);
let conversationId2 = "another conversation name";
let response2 = await bot.ask("Hello?", conversationId2);
console.log(response2);