1) Say hello (non‑streaming)
Send your first request with modelsmall-1 and get a complete response:
result.response[0].content.
2) Understand the response
The assistant’s text is inresult.response[0].content:
Want streaming output? See → Streaming Chat