Description
Describe the feature
Can the Agents SDK support OpenAI's realtime models like gpt-4o-mini-realtime-preview and gpt-4o-realtime-preview? It would be great to leverage these models for real-time responses within the agent framework.
Context
I came across #222 which discusses tracing and observability for the realtime API. Expanding support to include realtime models could improve both responsiveness and user experience for agents.
Use Case
Using realtime models would be especially useful for voice-based interactions where low latency is critical. Quick and natural responses are essential for maintaining smooth, human-like conversations in voice applications.
Would love to hear if this is on the roadmap or if there are any existing workarounds!
Thanks for the awesome work so far! 🙌