# Stream reasoning content as it arrives with OpenAI Responses

> We fixed handling of response.reasoning_summary_text.delta events in OpenAIResponses so that reasoning content is streamed incrementally as it is generated rather than being dropped or buffered.

- Published: 2026-04-10
- Author: Agno Team
- Category: Changelog
- Canonical: https://agno-com-nine.vercel.app/articles/stream-reasoning-content-as-it-arrives-with-openai-responses
- Markdown: https://agno-com-nine.vercel.app/articles/stream-reasoning-content-as-it-arrives-with-openai-responses.md

We fixed handling of `response.reasoning_summary_text.delta` events in `OpenAIResponses` so that reasoning content is streamed incrementally as it is generated rather than being dropped or buffered. Users now see the model's reasoning surface in real time alongside the response.