GPT-Live separates voice interaction from heavy reasoning—that's the real shift

别被全双工骗了：GPT-Live 真正改变的是你的语音编程模式

OpenAI launched GPT-Live on July 8, adding full-duplex and a delegation architecture to ChatGPT voice. Full-duplex lets you interrupt and talk while it works, but the principle isn't new—Moshi, Gemini Live, and ByteDance's Seeduplex all did it. The real change is delegation: the voice layer handles conversation while GPT-5.5 runs search, reasoning, and computation in parallel in the background, returning results as they arrive. This breaks the latency paradox where faster meant dumber. The voice model itself is limited—the System Card confirms it has no standalone tool access or code execution. No API yet; developers can only sign up for a waitlist. Realtime API remains the production workhorse at $64/M tokens for audio output. The post doesn't spell out whether custom tools can be plugged into the delegation layer or how much control developers will get over the black box.

Why it matters: OpenAI just shipped GPT-Live, and this analysis doesn't stop at full-duplex—it pulls out the delegation architecture as the real novelty. The author has technical judgment, laying out comparisons with Moshi, Gemini Live, and ByteDance's Seeduplex clearly. Points off because th...