Model API knowledge

[entrepeneur4lyf]

I know this isn't really an API issue... but, if I try to use any of the models to code an app using the API, I have to train it on the current API. Its knowledge is restricted to an old version. This wastes so much context and tokens it's crazy. I don't see a down side for OpenAI to train the models on the current API except that it gives the model self context. However, as a developer trying to use the model to code using the API it's extremely frustrating.