Ideas:
- When making a new chat with GPTs, allow users to choose which model to use (e.g. GPT-4, GPT-4o, GPT-4o mini, or o1)
- The ability to have an option in the model selector like “API Endpoint” that would allow developers familiar with the OpenAI API to choose a specific model to use.
- The ability to enable or disable real-time response streaming.
- The ability to switch between how real-time generated responses are shown during streaming (e.g. similar to the app or website)
- More UI customizability.