How to set temperature and other sampling parameters of model in Open AI Assistant api?

In the meantime, use the semantics of AssistantApi in the front and fulfill the request through chat completion in the back?

See Introducing openairetro -- AssistantAPI to ChatCompletion