Text Stream
DocsStream data as it's generated instead of waiting for the complete response. This is raw streaming: bytes flow through as they're ready, with no extra structure added. Perfect for simple LLM token streaming where you just want text to appear word-by-word. If you need typed events, message IDs, or automatic reconnection, see SSE streaming.
Prompt
What are AI agents and how do they work?
OpenAI/GPT-5 Nano
ReadyOutput will appear here as text streams in...
Reference Code
Loading...
Ready
Output will appear here...