Resolution criteria

This market will resolve to the date (in UTC) when OpenAI officially makes a GPT-Live-1 model (such as gpt-live-1 or gpt-live-1-mini) available for developer API use.

• Source of Truth: Official announcements published on the OpenAI Index / Blog, the OpenAI Developer Changelog, or via OpenAI's verified developer communications and official social media channels (e.g., @OpenAI or @OpenAIDevs on X).

• Release Phase: This market will resolve on the date of the first API release to developers, whether it is in private beta, limited preview, public beta, or general availability. If access is granted via a phased waitlist, it resolves to the date the first batch of external developers are granted active API access.

• Precondition/Edge Cases: If OpenAI releases the API under a modified name that is explicitly stated to be the developer API model for the GPT-Live-1 voice engine, this will satisfy the resolution criteria. If no such release occurs by September 1, 2026, the market will resolve to that date or N/A as determined by the market creator.

Background

OpenAI launched the GPT-Live-1 and GPT-Live-1 mini voice models on July 8, 2026. Built on a full-duplex architecture, GPT-Live allows for natural, real-time conversations where the model listens and speaks simultaneously, managing interruptions and pausing dynamically. At launch, the models were integrated directly into the ChatGPT Voice consumer experience, with GPT-Live-1 running by default for Go, Plus, and Pro tiers, and GPT-Live-1 mini powering the free tier.

While the consumer rollout is active, OpenAI has stated that developer API access is "planned soon". Developers can sign up for notifications regarding the API release on OpenAI's GPT-Live-1 in the API landing page.