Hi,
I’m just wondering if chatgpt is based on one unique model or a collection of model.
Specifically, when we are calling OpenAI API, we’re using a unique model, but chatgpt is a web tool. Does it use a collection of models (e.g., a set of models to filter harmful content, and if the content is not flagged, it is fed to the main model, e.g, gpt-4)?
Thank you.