Required prerequisites

• I have searched the Issue Tracker and Discussions that this hasn't already been reported. (+1 or comment there if it has.)
• Consider asking first in a Discussion.

Motivation

The current model module can use OpenAI's SDK to pass the stream parameter, but in the chat agent's step method, it actually waits for the model to completely finish responding before returning all the output at once. We want to achieve true streaming output in the chat agent.

Solution

No response

Alternatives

No response

Additional context

No response