
GPT-Live-1 is now available in the API. Bring ChatGPT’s natural back-and-forth to your app, with voice agents that listen while they speak and work with the models and harness you choose. https://t.c
The Hook
“Hey, what are we launching today? We're launching GPT-Live-1 in the API.”
Product Reveal
“It's a full duplex model for natural, expressive conversation, built to handle background noise and interruptions smoothly...”
Feature Teaser
“...while pairing with a backend model for tools and reasoning. Think of it like ChatGPT Voice... Wait, you were saying I don't have to worry about interruptions anymore?”
Feature Teaser
“Right, you can jump in, I can listen and speak at the same time. Sorry, there's some background noise here, can you still hear me okay? Yeah, you're coming through clearly. I can follow your voice even with noise around.”
Feature Teaser
“And you were saying something about a backend model? Mmhmm. GPT-Live-1 pairs with one for reasoning and tools. That's how I delegate actions to the robot and display here, while keeping the conversation flowing. In that case, can you say hi to everybody? Hi everyone! Great to be here. Welcome to the GPT-Live-1 launch.”
Call to Action
“And how much does this all cost? Of course. At just 5 cents per minute for the frontend model, GPT-Live-1 is priced for scale. That's the frontend price. Backend inference and tool services are separate. You can now bring conversational voice to production with GPT-Live-1 in the API today. Happy building.”
Synthesizable in Remotion · key components of the ai-voice-api-launch template.
Horizontally scrolling multi-lane audio waveform with synchronized event markers beneath audio channels.
Render SVG amplitude bars driven by current audio frame data, accompanied by linear timeline scrub cursor mapped to current frame.
Simulated low-res RGB LED grid displaying glowing cyan and yellow pixelated glyphs.
Grid of round div nodes modulated with CSS brightness and box-shadow based on a 2D boolean array state.
Purpose. Introduce the model naturally through dialogue rather than a corporate voiceover.
Execution. Presenter asks a question to an off-screen entity that turns out to be a desktop robot.
Purpose. Prove real-time interruption and background noise filtering claims live.
Execution. Background distraction introduced while user interrupts assistant mid-sentence.
Purpose. Remove developer adoption friction with clear pricing and API access.
Execution. LED frame displays rate prominently while presenter addresses viewer.