Seems like I am doing the opposite. I am building an OpenAI template right now for all of their biggest models. This will also include TTS and hopefully speech to text. Right now I just have to finish up GPT-4 with Vision so its ready for other people to use.