Making a real-time web app using assistants api without streaming in production is pain-full …
Messages under threads is essentially streaming. Consuming it directly from the endpoints is a question, How to listen the server sent events.
One of the HACK was to retrieve the latest assistant message from the msg_id, isn’t working…
In order to stream the new message OpenAI is likely going to need to add { stream: true } support to their GET /threads/{thread_id}/messages/{message_id} API.