OpenAI introduced GPT-Live-1 in the API on September 10. The model can listen and speak simultaneously and delegate deeper reasoning or actions to other models and tools. The announcement highlights interruption handling, configurable speaking style and telephony use cases.
Read the dated primary source ↓Why the separation matters
A natural conversation and a correct action are different engineering problems. A voice layer can keep an interaction moving while a backend works, but users still need to know whether a request is pending or complete. Smooth speech should not conceal a failed tool call.
What developers should test
Include pauses, corrections, background speech and ambiguous names in evaluation. Validate consequential information before acting and provide recovery when the connection drops. Compare complete task success and user control alongside the vendor’s reported voice benchmarks.
What this means in practice
Real-time APIs need to be evaluated as interactions rather than isolated response samples. Record connection behaviour, time to useful output and recovery from interruption. A published interface can enable new applications while still requiring the developer to manage permissions, state and the consequences of any connected action.
Questions readers ask
What should a real-time AI API pilot measure?
Measure useful response time, interruption handling, failures and task completion. Include the surrounding application, not only the model's output speed.
Follow the evidence.
Source-based coverage of the dated announcement below, with Promptwire analysis. Company-reported claims are attributed to their source. The update date reflects added context and answers; it does not change the original event date.
Public sources support the factual background. Analysis and illustrative exercises are identified in the text. How we assess AI · Editorial standards
Spot a factual error or an outdated detail?



