Skip to main content
GPT Live is a full-duplex voice model that talks to the caller directly while a backend text LLM runs your prompt, tools, and knowledge base. The voice model owns turn-taking and speech, so it answers fast; when it needs an action or information, it delegates to the backend and speaks the result in its own words. You set up GPT Live 1 (gpt-live-1) on a single-prompt voice agent, from the dashboard or through the API.

How it works

A GPT Live agent runs two models together:
  • Voice model (gpt-live-1) — hears and speaks to the caller, handles interruptions and turn-taking, and decides when it needs the backend. It follows the Retell LLM’s voice_prompt.
  • Backend text model — a cascading text LLM that runs the agent’s main prompt, tools, and knowledge base. It follows the Retell LLM’s general_prompt.
The voice model keeps the conversation flowing while the backend works, instead of pausing on a filler line. It never mentions the backend to the caller.

When to use it

Use GPT Live when you want speech-to-speech responsiveness but still need the tools, knowledge base, and prompt control of a text LLM. It fits single-prompt voice agents that take actions mid-call, such as booking an appointment or looking up an order, where a cascading model’s response gap is noticeable. It isn’t the right choice for a conversation flow, a multi-prompt agent, a transfer agent, or a chat agent. For those, use a cascading or speech-to-speech model instead.

Requirements

GPT Live works only on a single-prompt Retell LLM used by a voice agent, and it requires a paid account.
  • Single prompt only. A Retell LLM with states (a multi-prompt agent) is rejected.
  • Not for conversation flows or transfer agents. Setting type: "duplex" on a conversation flow, or on a transfer agent’s Retell LLM, is rejected.
  • A paid account. Like a speech-to-speech model, GPT Live isn’t available during the free trial.
  • No call-level override. An agent_override on a call can’t switch an agent to a duplex model.

Configure in the dashboard

On a single-prompt voice agent, GPT Live is a model you pick in the agent’s model selector, next to the cascading and speech-to-speech models.
1

Open a single-prompt voice agent

Open the agent you want, or create a new single-prompt voice agent. GPT Live isn’t offered on conversation flow, multi-prompt, transfer, or chat agents.
2

Select GPT Live 1

In the model selector, choose GPT Live 1, listed after the speech-to-speech models. Its price shows as $0.05/min + backend model.
3

Pick the backend model

Choose the Backend model next to the selector. It runs the agent’s main prompt, functions, and knowledge base when the voice model hands a request to the backend.
4

Write the conversation instructions

Fill in Conversation instructions, the voice model’s prompt (voice_prompt in the API). It covers how the agent talks and when to hand a request to the backend. A counter shows the token count against the 15,000-token limit. See Write the voice prompt.
5

Keep tasks in the main prompt

The agent’s main prompt is the backend’s prompt (general_prompt in the API): your task, tools, step-by-step procedures, and guardrails. Keep these there, not in the conversation instructions.
6

Choose an OpenAI preset voice

GPT Live supports OpenAI preset voices only. If the agent is on another voice, Retell switches it to an OpenAI preset when you select GPT Live 1. See Choose a voice.
The conversation instructions autosave as you edit. Retell keeps them when you switch the agent to another model, and uses them again if you switch back.

Configure with the API

Set model_config.type to duplex on the Retell LLM, and give the Retell LLM a voice_prompt: voice_prompt sits on the Retell LLM next to general_prompt, which is the backend’s prompt: your task, tools, and guardrails. Retell keeps the voice prompt when you switch the Retell LLM to another model, and uses it again if you switch back. Dynamic variables in the voice prompt are filled in once, when the call starts.
Attach the returned llm_id to a voice agent as its response engine, the same as any other Retell LLM. To change a duplex agent’s models later, send model_config on Update Retell LLM, and send voice_prompt to change the voice prompt. model_config.voice_prompt is rejected. The legacy flat model fields (model, s2s_model, model_temperature, and the rest) can’t describe a duplex config, so changing a duplex model’s models through them is rejected.

Write the voice prompt

The voice prompt tells GPT Live how to talk and when to hand a request to the backend. GPT Live hands off by generating a special token, <|SpawnThinking|>, so write your handoff rules with that exact phrase. Wording like “ask the backend” or “delegate” doesn’t trigger a handoff. See OpenAI’s GPT Live prompting guide for the full structure and more examples. Write the voice prompt in two parts:
  • Persona and speaking style. Who the agent is and how it talks, such as tone and reply length.
  • A delegation policy. Under Backend tools:, list what the backend can do, one capability per line. GPT Live uses this list to decide which requests to hand off. Then list when to hand off under Generate <|SpawnThinking|> when: and when to answer directly under Do not generate <|SpawnThinking|> when:. Keep these three labels exactly as written.
Keep tool definitions, step-by-step procedures, and facts such as prices in general_prompt and the knowledge base. GPT Live doesn’t see them: it asks the backend and says the answer in its own words. Retell adds two kinds of instructions to the voice model for you, so leave them out of the voice prompt:
  • Call control from the Retell LLM’s tools: ending the call, transfers, and keypad presses, including handing off when the caller says goodbye.
  • The agent’s settings, such as interruption sensitivity, backchannel, language, and guardrails. See What still applies.

Choose a voice

A GPT Live agent speaks in an OpenAI preset voice, set by the agent’s voice_id. The default is openai-Marin. GPT Live doesn’t support Retell custom voices or presets from other providers. Along with the OpenAI realtime presets it shares with speech-to-speech agents (such as openai-Marin and openai-Cedar), GPT Live can use an extra set of OpenAI presets that work only with gpt-live-1: Assigning one of these presets to a cascading or speech-to-speech agent is rejected with Voice <voice_id> is only supported with gpt-live-1. Assigning a voice GPT Live doesn’t support returns Voice <voice_id> is not supported for gpt-live-1. Use an OpenAI preset voice.

What still applies

Most agent settings carry over to GPT Live. The voice model honors the agent’s language, interruption sensitivity, responsiveness, backchannel, voice speed, pronunciation, boosted keywords, and guardrails, and it handles voicemail, IVR, do-not-call, and callback detection through the backend. The Welcome Message setting still decides who speaks first. Backchannel works a bit differently on GPT Live. With backchannel on, the voice model follows your frequency and word settings. With it off, GPT Live falls back to the voice model’s own default rather than going silent, so it may still add short acknowledgements while the caller speaks (as of November 2026). Keep backchannel on if you want to control how often it backchannels and which words it uses. For tools, Retell ignores each tool’s Speak during execution setting, since the voice model keeps talking to the caller while the backend runs, and it always speaks after a tool finishes. Some tools aren’t supported with GPT Live:
  • Agent swap — switching to another agent on the same call.
  • Warm (bridged) transfer and cancel transfer — a supervised transfer that bridges a third party, and cancelling one.
Saving a Retell LLM that uses one of these tools with gpt-live-1 is rejected. A plain call transfer, end call, and press digit all work.

Billing

GPT Live 1 costs $0.05 per minute of call time. The backend text LLM is billed separately for its own usage. Check the current Retell pricing before you deploy.