[NEW] The Most Complete OpenAI Plugin (Assistants, TTS, Chat stream...)

Hey everyone,
It’s been a while I started working on it and it has finally been accepted !

This plugin is the most complete OpenAI plugin for Bubble :

  • supports streaming for Text-to-Speech & Chat
  • supports Assistants
  • GPT-4, DALL-E 3, Text-to-speech…

It is a “bring your own key” plugin for security reasons.

Supported in the plugin

  • ChatCompletion with GPT-4, GPT-3.5 (turbo supported)
  • DALL-E 3
  • Text-to-Speech
  • Text-to-Speech Streaming
  • Transcription with Whisper
  • Assistant management (create, list, delete)
  • Thread management (create, delete, retrieve)
  • Messages management (create, retrieve)
  • Run Thread with Assistants
  • File management (create, retrieve, list, delete)
  • Moderation (create)
  • Embeddings (create)
  • Text completions streaming

Todo list (coming soon) :

  • Fine Tuning
  • Vision*

*Note :
You can already use Vision with the Chat Completions Action, it’s a bit trickier :

  • You have to set the model to : gpt-4-vision-preview
  • And to add images to message you will have to change a little bit the message object, example :
    { role: “user”, content: [
    {
    “type”: “text”,
    “text”: “What’s in this image?”
    },
    {
    “type”: “image_url”,
    “image_url”: {
    “url”: “image_url”
    }
    }
    ] }
    it has to be a string that can be parsed in JSON by the plugin.

The end user will have to input his own API key.

You can try it here with your own key (please try the TTS streaming it’s so good !)
https://vima-plugins-demo.bubbleapps.io/version-test/openai

And you can check how it works with the demo editor here : Vima-plugins-demo | Bubble Editor

And finally, the page of the plugin :

Please feel free to ask any questions regarding this plugin.

6 Likes

@Vincent_0x20100

Bookmarked. Streaming alone makes any plugin very worthwhile :smiley:

Great job!

2 Likes

[UPDATE] Fine-tuning is now integrated in the plugin !

You can now Fine-tune using the plugin :

  • Upload a file with fine-tune purpose
  • Get the ID of the file
  • Create a fine-tune using that file ID
  • You can also use all others fine-tune parameters to customize the fine-tune

Please see : https://platform.openai.com/docs/guides/fine-tuning
for more info concerning fine-tune and creating the good format for your fine-tune file.

Have fun with your fine-tunes !

1 Like