OpenAI has brought GPT-Live-1 to its API, giving developers access to the full-duplex voice model behind ChatGPT Voice. The release, announced on September 10, 2026, is priced at $0.05 per minute for the front-end voice layer.
What GPT-Live-1 Brings to the API
GPT-Live-1 is designed to listen and speak at the same time, a capability known as full-duplex voice. With the API release, that conversational system is no longer limited to ChatGPT Voice; it can now be integrated into third-party apps and business workflows.
How the Model Handles Reasoning and Actions
According to OpenAI, GPT-Live-1 delegates deeper reasoning and actions to the models and tools it is paired with. The company demonstrated this approach with Codex and ChatGPT Work, showing how the voice model can work alongside other systems rather than handling complex tasks on its own.
Pricing and Availability
The API release is priced at $0.05 per minute for the front-end voice layer. This makes the full-duplex voice model available to developers building conversational experiences, extending the technology beyond OpenAI’s own ChatGPT Voice product.
