For Developers/Models/GPT-Live-1.1
Closed SourceOpenAIReleased 2026-10

GPT-Live-1.1

OpenAI's updated full-duplex voice model for ChatGPT, with better accents and 13 new voices

Context window

Not published

Input / 1M tokens

Not announced

Output / 1M tokens

Not announced

Provider

OpenAI

Not in the OpenAI API yet, so there is no API price. In ChatGPT it is the default voice model on Plus and higher plans; other users get the smaller GPT-Live-1.1 Quip. · Data verified 2026-11-11

GPT-Live-1.1 (October 15, 2026) is the successor to GPT-Live-1, OpenAI's full-duplex voice model that listens and speaks at the same time. It is the default voice model for ChatGPT Plus and higher plans, including in Codex and Dots, while the smaller GPT-Live-1.1 Quip replaces GPT-Live-1 mini for other users. OpenAI says the update brings 13 new voices, considerably better performance in French, German, Hindi, Mexican Spanish, Korean and Japanese, more natural hand-offs to the backend reasoning model and better steerability. OpenAI says API access is coming soon; until then the API serves GPT-Live-1.

How to access it

ChatGPT Voice on all plans since October 15, 2026: GPT-Live-1.1 is the default on Plus and higher plans (also in Codex and Dots), and other users get GPT-Live-1.1 Quip. It is not in the OpenAI API yet, where gpt-live-1 is still the GPT-Live model; OpenAI says API access is coming soon.

Strengths

  • ✓Listens and speaks at the same time, with natural interruptions
  • ✓13 new ChatGPT voices
  • ✓Better accents in French, German, Hindi, Mexican Spanish, Korean and Japanese (OpenAI-reported)
  • ✓More natural delegation of hard questions to a reasoning model

Best for developers who...

Voice conversations in ChatGPTSpoken practice in the languages this update improvedResearch questions asked and answered by voice

When to choose it (and when not to)

Reach for GPT-Live-1.1 when...

  • →You use ChatGPT Voice and want the current model (the default on paid plans)
  • →You speak one of the languages OpenAI improved in this update

Look elsewhere if...

  • ✕You are building on the API today (use gpt-live-1 or gpt-realtime-2.1 until 1.1 reaches the API)
  • ✕You need published capability benchmarks: OpenAI has released only safety evaluations for 1.1

How to use it

  • ›Interrupt it mid-answer the same way you would a person
  • ›Ask it to look something up or work through a problem; it hands that work to a backend model while the conversation continues
  • ›Try the new voices in ChatGPT settings

Tools powered by GPT-Live-1.1

Compare GPT-Live-1.1

Compare GPT-Live-1.1 with any other model

Build a comparison →
All model comparisons →

Learn the concepts