The new voice models can listen and speak at the same time — and hand a question to a frontier model like GPT-5.5 for a live web search mid-conversation. GPT-Live-1 is now the default ChatGPT Voice model for paid users.
2026-07-08 · by Moe Ameen
OpenAI began rolling out GPT-Live on July 8, 2026, a new generation of voice models that power ChatGPT Voice. The core change is that they are full-duplex: the model can take in audio and produce speech at the same time. In practice, that means it can drop in a quick "mhmm" or "got it," trade fast back-and-forth, interrupt less, and stay quiet when you pause to think — a conversation that feels closer to talking to a person than to issuing commands and waiting for a reply.
Two versions shipped together. GPT-Live-1 becomes the default model behind ChatGPT Voice for Go, Plus, and Pro subscribers, while a smaller GPT-Live-1 mini becomes the default for free users. Both reach ChatGPT users worldwide across iOS, Android, and the web. OpenAI said more than 150 million people already talk to ChatGPT every week through its Voice and Dictation features — the install base this upgrade lands on.
The other notable piece is that voice is no longer isolated from search and reasoning. Mid-conversation, GPT-Live can hand a question to a frontier model such as GPT-5.5 for a web search or heavier reasoning, then keep talking, and it can surface visual cards for things like weather, stocks, and sports while you speak. Users can set the reasoning level — Instant for fast replies, or Medium and High for questions worth more thinking time — and the model keeps support for search, memory, images, and file uploads. Video and screen sharing are not part of this release; OpenAI said both are being worked on. Treat exact plan details and availability as a fast-moving snapshot and confirm them on OpenAI's own pages.
There are two ways to act on this today, and both run through Kompozy. The first is the story itself: a natural, search-capable OpenAI voice is a subject your audience is searching for this week. With Kompozy you turn "OpenAI launched GPT-Live, and here's what full-duplex voice changes" into an explainer Blog Article, a comparison Carousel against other AI voice assistants, native Text Posts for X and LinkedIn, and a short — the whole set drafted in your voice via the Persona Brief and scheduled across all nine social platforms plus blog and email from one queue, while search interest is spiking.
The second is the workflow the launch actually unlocks. GPT-Live is a fast, hands-free way to say the idea out loud and let it web-search to sharpen the facts — but it hands you a transcript, not a post. That's the exact seam Kompozy closes. Capture the conversation, paste it into Quick Ingest, and one voice note fans out into a Persona Short where your face-locked avatar reads the script, a brand-exact carousel through HyperFrames, quote graphics, a newsletter, and text posts — then publishes the set everywhere from a single queue with Autopilot and a per-post review pipeline. Talk the idea into GPT-Live between takes; let Kompozy turn it into the week's on-brand content across every platform the voice model can't reach.
GPT-Live is a new generation of OpenAI voice models that power ChatGPT Voice, rolled out starting July 8, 2026. They are full-duplex — they can listen and speak at the same time — so conversations feel more natural, and mid-conversation the model can hand a question to a frontier model like GPT-5.5 for a live web search or deeper reasoning.
GPT-Live-1 is the default ChatGPT Voice model for Go, Plus, and Pro users, and GPT-Live-1 mini is the default for free users. It rolled out to ChatGPT users worldwide on iOS, Android, and the web. Access comes with your ChatGPT plan — there is no separate GPT-Live price.
Not in this release. GPT-Live launched as a voice upgrade — full-duplex conversation plus in-call web search — and OpenAI said voice with video and screen sharing in ChatGPT are being worked on but are not part of the initial rollout.
Use it as a hands-free ideation and research partner — talk through an angle and let it web-search to sharpen the facts — then capture the transcript and run it through a content engine like Kompozy, which turns the idea into a blog, carousel, persona video, text posts, and a newsletter and publishes them across nine platforms.