OpenAI module

The openai module allows you to communicate with the OpenAI API directly from machine code.

Importing the module

const OpenAI = require('openai')

Client initialization

const openai = new OpenAI({ apiKey: 'your_api_key' })

Chat completions (conversational AI)

const response = await openai.chat.completions.create({
  model: 'gpt-3.5-turbo',
  messages: [
    { role: 'system', content: 'You are a helpful assistant.' },
    { role: 'user', content: 'Hello, what can you do?' }
  ],
  temperature: 0.7,
  max_tokens: 100
})
const answer = response.choices[0].message.content

Text completions (classic prompt)

const response = await openai.completions.create({
  model: 'text-davinci-003',
  prompt: 'Write a short story about a cat.',
  temperature: 0.7,
  max_tokens: 100
})
const text = response.choices[0].text

Image generation

const response = await openai.images.generate({
  prompt: 'A cat riding a bike in Prague',
  n: 1,
  size: '512x512'
})
const imageUrl = response.data[0].url

Example: Sending an image to OpenAI API using blobToBase64

If you need to send an image (e.g. for vision models), you can convert a BlobReference to a base64 string and include it in your API request:

// Assume you have a BlobReference 'imageBlob' and a valid machineContext
const base64Image = await blobToBase64(machineContext, imageBlob)

const response = await openai.chat.completions.create({
  model: 'gpt-4-vision-preview',
  messages: [
    {
      role: 'user',
      content: [
        { type: 'text', content: 'Describe this image.' },
        { type: 'image_url', image_url: { "url": `data:image/png;base64,${base64Image}` } }
      ]
    }
  ]
})
const answer = response.choices[0].message.content

Replace image/png with the correct MIME type if needed. This example shows how to send an image as base64 to OpenAI's vision model.

Parameters

• model: Model name (e.g. 'gpt-3.5-turbo', 'text-davinci-003')
• messages: Array of messages for chat (role: 'system', 'user', 'assistant')
• prompt: Text prompt for text generation
• temperature: Response creativity (0-2)
• max_tokens: Maximum response length
• More parameters: see OpenAI API documentation

Notes

• You get your API key in the OpenAI dashboard.
• The module is available only in sandbox environments with OpenAI access enabled.
• Results may be billed according to your OpenAI plan.