OpenAI Launches GPT-Live-1

OpenAI has launched GPT-Live-1, giving developers a natural voice model for building voice-enabled apps and business workflows.

First introduced in ChatGPT, GPT-Live-1 can listen and speak at the same time, and can delegate deeper reasoning and actions to the models and tools with which it is paired.

With GPT-Live-1, developers can steer and customize voice experiences around their users, workflows, and goals.,/p>

Key strengths of GPT-Live-1 in the API include the following:

• Interruption handling via a single model that reasons over incoming and outgoing audio together.
• Reasoning and tool calling delegation to a back-end text model like GPT-6 Astra or a third-party model.
• The ability toshape agents' tone, pace, and conversational style through the system prompt.
• Silent context management and background noise handling.
• Improved context retention and conversational quality across extended interactions.
• Deployment of full-duplex voice agents for phone calls, from restaurant reservations to customer support.

Developers choose the models, tools, and agent harness behind the conversation.

GPT-Live-1 natively provides transcripts and response text. It also offers strong alphanumeric understanding and supports keyword biasing and turn detection.

With GPT-Live-1, OpenAI is expanding from a small set of real-time voices to a broader selection across accents, dialects, and languages.

Another way to build voice workflows on top of GPT-Live-1 is with OpenAI Presence, which uses the model to power real-time voice interactions. Presence helps enterprises deploy trusted AI agents that can answer questions, resolve issues, use company systems, take approved actions, and escalate to people when needed.

One of the first integration partners for GPT-Live-1 is Twilio, which has created a link between Twilio Agent Connect and GPT-Live-1. Twilio's software development kit connects OpenAI's voice models directly to its own Voice and Conversations layer without requiring custom turn-taking, audio streaming, interruption handling, or cross-channel state tracking. Acting as middleware between large language model applications and Twilio's platform, Agent Connect links OpenAI's voice model directly to voice, messaging, and Conversation Memory. OpenAI provides advanced real-time speech-to-speech intelligence in GPT-Live-1, while Twilio brings the trusted communications platform, global voice carrier network, and scale needed to run customer conversations in production. Agent Connect lets users integrate real-time AI in just a few configuration steps and deliver personalized, context-aware customer experiences that remember history across every channel.