The Streaming API allows you to receive real-time updates as Wren AI processes your request — from understanding the question, to generating SQL, running it, and optionally returning summaries.

These endpoints use Server-Sent Events (SSE) to deliver a continuous stream of structured events. This is ideal for interactive applications such as chat interfaces, notebooks, or dashboards where users benefit from step-by-step feedback.

Use Cases

✅ Display live reasoning and SQL generation feedback in chatbots
✅ Show SQL execution progress in real time
✅ Render streaming text response
✅ Improve perceived latency by showing intermediate steps

Available Streaming Endpoints

Endpoint	Description
`POST /stream/ask`	Full pipeline: generates SQL, executes it, and returns a summary
`POST /stream/generate_sql`	Only generates the SQL based on a natural language question