Justin Uberti

Today, we're launching our third-gen voice model and architecture, GPT-Live. GPT-Live is a full-duplex model with built-in async delegation, which allows it to deliver incredibly natural conversation along with the intelligence you expect from ChatGPT. openai.com/index/introduc…

@iandanforth crawl then walk then run

@juberti Then (and I'm also hating on this, lets be clear) I can't use live as an orchestrating control / interface to the whole app (desktop/mobile/whatever). I want to just ask it to start sessions check in on them etc.

Converting the haters, one by one 😊

@juberti Grrr am hater. Please add a view of the text output at the same time in mobile. I need to be able to read the romaji of the vocab its explaining to me. Grrr very angry, convert me!

@wwd7086 > Speech, video, generated interfaces, tools, and robot actions may ultimately become different output streams of the same kind of real-time interactive model. 👍

What interests me about ChatGPT Live isn’t just that voice interaction feels more natural. It points toward a very different kind of personal computer. Imagine a wearable device that can hear what you hear, see what you see, and remain available throughout the day. You wouldn’t need to stop, open an app, and carefully formulate a prompt. You could simply speak whenever you wanted, while the assistant continuously understood the surrounding context. ChatGPT Live doesn’t create that entire experience yet, but it makes the interaction layer feel surprisingly close. I would genuinely want a device like this, assuming the privacy, trust, battery life, and user-control problems were handled well. There is also a broader convergence happening across AI research: • Add continuous video input, and the model can understand what you are seeing. • Add visual output, and it can have a generated physical appearance, present information spatially, or create interfaces on the fly. • Add “action” as another output modality, and the model can begin affecting the physical world through a robot. At that point, the boundary between a personal AI assistant and a robotics foundation model starts to blur. Both continuously perceive the world, maintain state over time, decide when to respond, and produce actions that influence what happens next. Speech, video, generated interfaces, tools, and robot actions may ultimately become different output streams of the same kind of real-time interactive model.

@dpetrou Thanks for the report. Some of our engines got a bit overloaded this afternoon. Are you still seeing this?

@juberti, the "live" chatgpt feature is glitching bad -- sounds like dropped frames galore. both on my fast wifi and many-barred 5G. yesterday and today. NYC.

@simpforasi no. gpt-live is a whole new architecture, and has given us a much higher ceiling

@juberti Will the time between live-1 and live-next be the same as between AVM and live-1 ?

@muratcan it just wants to be sure

@juberti Is this a really small model? It’s always “checking” whenever I ask a knowledge question.

@signulll Wow, thanks. Genuinely unaccustomed to these sorts of sentiments on X.

okay, enough openai criticism for a second (still gonna give product feedback when it’s deserved). it’s been a little too easy. i’ve been using gpt live a lot, like a lot lot, & it’s genuinely fucking amazing. the feature to let the model speak like a human “hmm”, “let me check”, brief pauses, & natural turn taking while calling a bigger model or searching or using tools in the background is a fantastic execution. almost all of my conversations have been nothing but pleasant, i can’t describe the feeling but it’s i think a little too good at this point. nothing else is particularly close in this category right now at all. anthropic’s voice experience still feels very very underbaked (after thought really), & gemini’s personality & intelligence seem lacking. & since it’s already reaching an enormous user base, every great gpt live interaction makes people more confident in ai as a product category instead of walking away thinking ai is a gimmick. really great work.

GPT-Live is now fully rolled out to all ChatGPT users globally. We're also doubling everyone's voice usage limit for the whole weekend so you can try more of it. Have fun! 🎉

We've improved memory reliability in GPT-Live. Let us know how it feels now!

@huskirl good old boots and cats

Is this good music?

Now that GPT-Live is out, tonight I'm gonna party like it's 1995

The new ChatGPT voice is quite impressive to use, really worth a minute to try it out on your phone. (while staying aware that the voice model is not going to be as smart as a full thinking model)

Impressive use of GPT-Live's translation capability during the GPT-5.6 livestream today (direct link) youtube.com/live/Wq45rvPGN…

it’s been a privilege to help shape the research alongside such a thoughtful, talented, and relentless group of people. Huge credits to @ritheshkumar_ , Yuchen, Ehsan and rest of the team who brought this model to life. Super proud. Additionally, it has been super fun to collaborate with @athyuttamre @ding_ran and @juberti and their teams to get the end-to-end experience really great!!

@Babybat_fangs @AgorithmAg Could you share an example?

@juberti @AgorithmAg Its not good, I am just being honest, its lacking the story telling. Please l beg you keep standard ! I feel so heartbroken with this update.

Today, we're launching our third-gen voice model and architecture, GPT-Live. GPT-Live is a full-duplex model with built-in async delegation, which allows it to deliver incredibly natural conversation along with the intelligence you expect from ChatGPT. openai.com/index/introduc…

Huh, I thought Snow Crash was fiction.

@RachaelMillett2 @AgorithmAg welp, ok, that's pretty clear feedback

@juberti @AgorithmAg It’s just a no. It doesn’t even sound like cove, not to mention it sounds like a customer service rep. All you had to do was make standard voice mode bidirectional, whole task right there. Advanced voice flopped and apparently zero things were learnt