GPT4js 🔮

GPT4js is a package that simplifies interaction with various AI models, eliminating the need for an API Key or any other authorization method to access these chat completions and image generation models.

This package can be used in Node.js or Browser environments.

📚 Table of Contents

• GPT4js 🔮
  • 📚 Table of Contents
  • 🛠️ Installation
    • Using NPM
    • Using Yarn
    • Using Bun
• 🧩 Examples
  • 📤 Chat Completion
    • ⚙️ Basic Usage
      • Simple Fetch
      • Give Your Instructions
      • Conversation Roles
    • 🔩 Configurable Options
    • 🚀 Chat Completion Providers
    • 📚 Chat Completion Models
• 📷 Image Generation
  • 📹 Example usage
  • 🌐 Image Generation Provider Options
  • 🧮 Number Type Options
  • 🖼️ Image Generation Providers
  • 🧠 Google Chrome AI
    • Setting Browser
      • Simple Usage
• 🧪 Testing
• 🚧 Building
  • Webpack
  • Bun
• 🤝 Contribute

🛠️ Installation

Using NPM

npm install gpt4js

Using Yarn

yarn add gpt4js

Using Bun

bun add gpt4js

🧩 Examples

📤 Chat Completion

With the chatCompletion function, you can obtain a textual response to a conversation with some context, using providers and models designed for this task. Additionally, you can manipulate the answer before converting it to a stream or force the AI to give you a certain answer by generating several retries.

⚙️ Basic Usage

Simple Fetch

It will capture the messages and the context, and any provider will respond with a string.

// CommonJS
const getGPT4js = require("gpt4js");
const GPT4js = await getGPT4js();
// ESM
import GPT4js from "gpt4js";

const messages = [{ role: "user", content: "hi!" }];
const options = {
  provider: "Nextway",
  model: "gpt-4o-free",
};

(async () => {
  const provider = GPT4js.createProvider(options.provider);
  try {
    const text = await provider.chatCompletion(messages, options, (data) => {
      console.log(data);
    });
    console.log(text);
  } catch (error) {
    console.error("Error:", error);
  }
})();

Note: The conversation needs to include at least one message with the role user to provide a proper answer.

Give Your Instructions

You can provide your own instructions for the conversation before it starts using the system role.

const messages = [
  { role: "system", content: "You're an expert bot in programming." },
  { role: "user", content: "Hi, write me something." },
];
const options = {
  provider: "Nextway",
  model: "gpt-4o-free",
};

(async () => {
  const provider = GPT4js.createProvider(options.provider);
  try {
    const text = await provider.chatCompletion(messages, options, (data) => {
      console.log(data);
    });
    console.log(text);
  } catch (error) {
    console.error("Error:", error);
  }
})();

Conversation Roles

Role	Description
system	Used for providing instructions and context prior to the conversation.
user	Used to identify user messages
assistant	Used to identify AI messages

🔩 Configurable Options

Option	Type	Description
provider	string	Choose the provider to use for chat completions. Possible values include Nextway, BlackBox, etc. This determines which service will handle the request.
model	string	Choose the model to use by a provider that supports it. For example, gpt-4, gpt-3.5-turbo, etc. This specifies the particular language model for generating completions.
stream	boolean	Determine if the data should be streamed in parts or not. If true, the response will be streamed in real-time as it's generated. If false, the response will be sent all at once.
temperature	number	Set the temperature to control the randomness of the output. A value between 0 and 1 where higher values (closer to 1) make the output more random, and lower values (closer to 0) make it more deterministic.
webSearch	boolean	Enable or disable web search functionality. If true, the system can perform web searches to gather real-time information. If false, it relies solely on pre-existing data.
codeModelMode	boolean	Enable or disable the code model mode. If true, the system will use a model optimized for understanding and generating code. If false, it uses the general-purpose language model.
isChromeExt	boolean	Specify whether the system is being used as a Chrome extension. If true, it indicates integration with Chrome, possibly affecting certain functionalities and permissions.

🚀 Chat Completion Providers

Website	Provider	GPT-3.5	GPT-4	Stream	Status
Aryahcr	Aryahcr	✔️	✔️	✔️
BlackBox	BlackBox	❌	❌	❌
Nextway	Nextway	✔️	✔️	✔️
Chrome	Chrome	❌	❌	✔️
Ollama	Ollama	❌	❌	✔️
Alibaba	Alibaba	✔️	❌	✔️
ChatBotRu	ChatBotRu	❌	✔️	✔️

📚 Chat Completion Models

Model	Providers that support it
gpt-4	Aryahcr, Nextway, ChatBotRu
gpt-4-0613	Aryahcr
gpt-4-32k	Aryahcr
gpt-4-0314	Aryahcr
gpt-4o-free	Nextway
gpt-4o	ChatBotRu
gpt-4-32k-0314	Aryahcr
gpt-4-turbo	ChatBotRu
gpt-3.5-turbo	Aryahcr, Nextway, Alibaba
gpt-3.5-turbo-16k	Aryahcr
gpt-3.5-turbo-0613	Aryahcr
gpt-3.5-turbo-16k-0613	Aryahcr
gpt-3.5-turbo-0301	Aryahcr
text-davinci-003	Aryahcr
text-davinci-002	Aryahcr
code-davinci-002	Aryahcr
gpt-3	Aryahcr
text-curie-001	Aryahcr
text-babbage-001	Aryahcr
text-ada-001	Aryahcr
davinci	Aryahcr
curie	Aryahcr
babbage	Aryahcr
ada	Aryahcr
babbage-002	Aryahcr
davinci-002	Aryahcr
gemini-pro	Nextway
gemini-nano	Chrome
All Ollama models	Ollama
SparkDesk-v1.1	Alibaba
deepseek-coder	Alibaba
deepseek-chat	Alibaba
Qwen2-7B-Instruct	Alibaba
glm4-9B-chat	Alibaba
chatglm3-6B	Alibaba
Yi-1.5-9B-Chat	Alibaba
llama-3.1-405b-instruct-free	Nextway

📷 Image Generation

📹 Example usage

const options = {
  provider: "DALLE2",
};

(async () => {
  const provider = GPT4js.createProvider(options.provider);
  try {
    const base64 = await provider.imageGeneration("wood", options);
    console.log(base64);
  } catch (error) {
    console.error("Error:", error);
  }
})();

With the imageGeneration function, you can generate images from textual input along with optional parameters to customize and stylize the images in various artistic styles.

🌐 Image Generation Provider Options

Option	Type	Description
negativePrompt	string	Indicates the direction not to take in production.
height	number	Specifies the image height.
width	number	Specifies the image width.
samplingSteps	number	Specifies the number of iterations. A higher number results in higher quality.
samplingMethod	string	Selects a sampling method to control the diversity, quality, and coherence of images.
cfgScale	number	Specifies the Classifier-Free Guidance to control how closely the generated image adheres to the given text prompt.

🧮 Number Type Options

Provider	Supported Number Type Options and Values
StableDiffusion	- height: Default 512, Min 50, Max 1024 - width: Default 512, Min 50, Max 1024 - samplingSteps: Default 25, Min 1, Max 30 - cfgScale: Default 7, Min 1, Max 20

🖼️ Image Generation Providers

Provider	Status	Default Style
Dalle2		Semi-realistic, detailed with vivid colors and natural lighting.
StableDiffusion		Photorealistic, capturing fine details and textures to simulate real-life scenes.

🧠 Google Chrome AI

Warning: This is an experimental feature and may not work correctly, it only works in Google Chrome 127 or higher (Chrome Dev). Also history is not supported

Setting Browser

1. chrome://flags/#prompt-api-for-gemini-nano Select 'Enabled'

2. chrome://flags/#optimization-guide-on-device-model Select 'Enabled BypassPrefRequirement'

3. chrome://components Click 'Check for Update' on Optimization Guide On Device Model to download the model. If you don't see Optimization Guide, ensure you have set the flags correctly above, relaunch your browser, and refresh the page.

Simple Usage

const messages = [{ role: "user", content: "hi!" }];
const options = {
  provider: "Chrome",
};

(async () => {
  const provider = GPT4js.createProvider(options.provider);
  try {
    const text = await provider.chatCompletion(messages, options, (data) => {
      console.log(data);
    });
    console.log(text);
  } catch (error) {
    console.error("Error:", error);
  }
})();

🧪 Testing

Running: npm test

🚧 Building

Webpack

• npm run build - Build using Webpack.
• npm run dev - Live development build with Webpack.

Bun

• npm run build:bun - Build using Bun.
• npm run dev:bun - Live development build with Bun.

🤝 Contribute

If you'd like to contribute to this project, you can do so directly on GitHub. Additionally, if you encounter any errors that hinder your use of any project functionality, please report them here. Your feedback helps our community access AI tools freely!

---