Robots Atlas>ROBOTS ATLAS
GPT-Live-1 mini
AI Modelsโ€บGPT

GPT-Live-1 mini

1 miniย ยทย Family: GPT
A lighter variant of OpenAI's conversational voice model released on 8 July 2026, built on a full-duplex architecture. It becomes the default voice model for users on ChatGPT's free tier.
โœ“ Activeโœ“ Public accessAudioMultimodal๐Ÿ“ GPT
Release date
8 July 2026
๐ŸขOpenAIProducer
Access:HostedDeployment:โ˜ Cloud

Overview

GPT-Live-1 mini is a lighter variant of OpenAI's conversational voice model, released on 8 July 2026 alongside GPT-Live-1. It shares the same full-duplex architecture โ€” listening and speaking simultaneously, removing the rigid turn-taking rhythm of Advanced Voice Mode. It becomes the default voice model for users on ChatGPT's free tier (the full GPT-Live-1 serves the Go, Plus and Pro tiers). Like the full version, it delegates queries requiring search or reasoning to GPT-5.5 running in the background.

Classification
AudioMultimodal
Family: GPT
Access & deployment
Hosted
Cloud
Weights: Closed
Key parameters
โœ“ Tools
๐Ÿ“ฅ Input: text, audio

Technical specification

License
Proprietary
Features:โœ“ Tool use
Modalities
โฌ‡ Input
textaudio
โฌ† Output
textaudio

Capabilities and applications

Native model capabilities
Voice Conversation
Ability to conduct multi-turn real-time voice conversations with context retention and natural speech pacing.
Category: speech
Speech to text
Category: speech
Text to speech
Category: speech
Real-time inference
The model's ability to generate responses with very low latency (>1000 tokens/sec) on specialized inference hardware (e.g. Cerebras WSE), enabling interactive, turn-by-turn collaboration with a human.
Category: coding
Multilingual
Understanding and generating text in many languages.
Category: language

Sources and related pages