OpenAI patents developer tools for testing real-time AI conversations

1. A method of interacting with real-time generative response engines having one or more user interface (UI) elements, the method comprising: presenting an interaction control interface for use with a real-time generative response engine; presenting a start session option within the interaction control interface; and responsive to receiving a user input selecting the start session option, presenting, in the interaction control interface, a generation control UI element that is effective to toggle between a generating state corresponding to the real-time generative response engine generating a response and a non-generating state corresponding to the real-time generative response engine not generating the response, wherein the interaction control interface is configured to: receive a first input that is effective to start a real-time communication session, wherein the first input comprises the user input selecting the start session option, receive a second input that is effective to stop or pause the real-time communication session, receive a third input that is effective to cause the real-time generative response engine to stop generating a particular response, wherein the third input is effective to cause the real-time generative response engine to suspend an active response generation operation for the particular response while the response is being streamed without terminating the real-time communication session, and receive a fourth input comprising a subsequent generation command that is effective to cause the real-time generative response engine to initiate generation of a response based on a chat history of the real-time communication session, wherein toggling the generation control UI element to the generating state causes a first command to be sent to the real-time generative response engine to stream the response based on a chat history of the session, and wherein toggling the generation control UI element to the not generating state causes a second command to be sent to the real-time generative response engine to surrender a turn and discontinue streaming the response while maintaining the session.