Urgent.News

What's breaking now, across thousands of outlets.

AI

OpenAI Just Put GPT-Live-1.1 in the API, and Idle Voice Sessions Are Now 60% Cheaper

Voice agents have had an expensive habit: they charge you full price while nobody is talking. A customer puts the phone down for a minute, the agent sits there listening to silence, and the meter keeps running at the active rate. OpenAI's answer, announced 19 November 2026 and in the API from 20 November, is twofold: a better model, GPT-Live-1.1, and a Standby mode that cuts the idle rate by 60%.…

OpenAI has introduced two major updates to its voice agent service, GPT-Live-1.1. First, a new model called GPT-Live-1.1 has been made available in the OpenAI API. This model offers improved transcription accuracy, more natural and consistent accents, better handling of precise details, and better instruction following compared to its predecessor, GPT-Live-1.

Second, OpenAI has introduced a Standby mode that reduces the cost of idle voice sessions by 60%. Under this mode, the model pauses inference when there is a period of silence, and billing drops from $0.05 to $0.02 per active/idle minute. This change can lead to significant cost savings for voice agent services. For example, a voice agent working an eight-hour shift with 30% talking time and 70% idle time would see a roughly 40% reduction in daily costs.

The Standby mode is available for both GPT-Live-1 and GPT-Live-1.1, so migration to the newer model is not required to benefit from the idle rate reduction. Additionally, GPT-Live-1.1 allows delegation of complex tasks to a backend model, enabling more efficient use of resources and better handling of structured outputs.

Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.

Read the original at dev.to →

More in AI

More from Friday 20 November →