AMA on the 17th of December with OpenAI's API Team: Post Your Questions Here

Can there be an easier way of doing continuation (with structured outputs and streaming)? Right now it’s rather difficult to stitch 2 assistant responses appropriately