Getting an API key

1. Join the Discord server.
2. Get your API key from the #Bot channel by using the /key command.
3. Copy the API key that the bot gives you

Note: This is not an OpenAI API key.

Installation

1. Run npm init -y
2. Go to your package.json file and add "type": "module" at the end (before the } )
3. Run npm i gpt-proxy
4. Add import { gpt } from 'chatgpt-api-free' to your index.js file

Note: Step 2 will make you not able to use require() instead you will have to switch to import/export so try to start new projects with this library instead of adding it to big projects

Creating gpt instance

const model = new gpt({
    'api_key': 'your API key',
    'temperature': 0.7, 
    'max_tokens': 256 // max: 4090
})

// new gpt(<config: <api_key: string> <temperature: number> <max_tokens: number> >)

Text Completion

const completion = await model.text_completion('Human: Hello!\nAI:', [
    'Human:',
    'AI:'
])


console.log(completion.choices[0].text)

// text_completion(<prompt: string>, <stop: array>)

Chat Completion

const completion = await model.chat_completion([
        {
            'role': 'system',
            'content': `You are a helpful assistant, your name is Joe`

        },
        {
            'role': 'user',
            'content': 'What\'s your name?'
        }
    ])

    console.log(completion.choices[0].message.content)

// text_completion(<prompt: string>, <stop: array>)

Have fun using this library, its great for small projects like chat bots, you can support the project that this library uses in the discord server

Coming soon:

Whisper model

Fine tuning (maybe)