Feature Request: Support for Multiple Completions in OpenAI API Integration

[simonorzel26]

Feature Description

I am interested in expanding the functionality by incorporating multiple completions (i.e., n: 3) from the OpenAI API into a single request. This feature will allow a variety of response suggestions to the user, enhancing their experience by providing multiple dialogue options or content variations from which they can choose.

const response = await openai.createCompletion({
  model: 'gpt-3.5-turbo',
  prompt: 'Your prompt here',
  max_tokens: 150,
  n: 3,  // Requesting three completions
});

Use Case

This adjustment will allow applications to process and render multiple AI-generated responses simultaneously, offering a richer and more interactive experience for suggestive and multiple completions. For instance feedback loops for messages that have better intended use. (One message is less descriptive than the other, so it is saved as history and used in the future for multiple shot prompting)

Additional context

Currently using the render() method, we are very limited to the use.

render({ model: 'gpt-3.5-turbo', provider: openai, initial: <SpinnerMessage />, messages: [