OpenAI's GPT-Live API streams slower than realtime; new continuously-generating architecture blamed

juberti · x · 2026-09-19

Users report the GPT-Live API streams audio slower than realtime, unlike earlier OpenAI realtime models. juberti explains it's a whole new architecture that generates output continuously in real time and says WebRTC/SIP endpoints will offer the best performance.

Original post →

More from Models

Models channel →