Story
Loading the article brief, supporting context, and related editorial blocks.
AI BriefWire / Briefing
OpenAI has launched the Realtime API to enable instant interaction with AI models. This API allows developers to build applications that respond immediately to user inputs. It matters because it enhances user experience by reducing latency in AI-powered applications.
%22%2F%3E%0A%3Ccircle%20cx%3D%22978%22%20cy%3D%22150%22%20r%3D%22210%22%20fill%3D%22rgba(124%2C166%2C199%2C0.16)%22%2F%3E%0A%3Ccircle%20cx%3D%22978%22%20cy%3D%22150%22%20r%3D%22122%22%20fill%3D%22rgba(255%2C255%2C255%2C0.04)%22%20stroke%3D%22rgba(255%2C255%2C255%2C0.12)%22%2F%3E%0A%3Cpath%20d%3D%22M70%20428H1130%22%20stroke%3D%22rgba(208%2C122%2C85%2C0.28)%22%20stroke-width%3D%222%22%20stroke-dasharray%3D%2210%2014%22%2F%3E%0A%3Crect%20x%3D%2276%22%20y%3D%2272%22%20width%3D%22360%22%20height%3D%2242%22%20rx%3D%2221%22%20fill%3D%22rgba(255%2C255%2C255%2C0.06)%22%20stroke%3D%22rgba(255%2C255%2C255%2C0.1)%22%2F%3E%0A%3Ctext%20x%3D%2298%22%20y%3D%2299%22%20fill%3D%22rgba(242%2C244%2C247%2C0.82)%22%20font-family%3D%22Arial%2C%20Helvetica%2C%20sans-serif%22%20font-size%3D%2217%22%20font-weight%3D%22700%22%20letter-spacing%3D%223%22%3EOPENAI%20NEWS%3C%2Ftext%3E%0A%3Ctext%20x%3D%2278%22%20y%3D%22212%22%20fill%3D%22%23F2F4F7%22%20font-family%3D%22Georgia%2C%20'Times%20New%20Roman'%2C%20serif%22%20font-size%3D%2254%22%20font-weight%3D%22700%22%3E%3Ctspan%20x%3D%2278%22%20dy%3D%220%22%3EIntroducing%20the%20Realtime%3C%2Ftspan%3E%3Ctspan%20x%3D%2278%22%20dy%3D%2262%22%3EAPI%3C%2Ftspan%3E%3C%2Ftext%3E%0A%3Crect%20x%3D%2278%22%20y%3D%22502%22%20width%3D%22260%22%20height%3D%2246%22%20rx%3D%2223%22%20fill%3D%22rgba(208%2C122%2C85%2C0.18)%22%20stroke%3D%22rgba(208%2C122%2C85%2C0.32)%22%2F%3E%0A%3Ctext%20x%3D%22104%22%20y%3D%22532%22%20fill%3D%22%23F2F4F7%22%20font-family%3D%22Arial%2C%20Helvetica%2C%20sans-serif%22%20font-size%3D%2222%22%20font-weight%3D%22700%22%3ECore%20AI%3C%2Ftext%3E%0A%3Crect%20x%3D%22900%22%20y%3D%22410%22%20width%3D%22190%22%20height%3D%22190%22%20rx%3D%2238%22%20fill%3D%22rgba(255%2C255%2C255%2C0.06)%22%20stroke%3D%22rgba(255%2C255%2C255%2C0.12)%22%2F%3E%0A%3Ctext%20x%3D%22995%22%20y%3D%22522%22%20text-anchor%3D%22middle%22%20fill%3D%22%23F2F4F7%22%20font-family%3D%22Arial%2C%20Helvetica%2C%20sans-serif%22%20font-size%3D%2292%22%20font-weight%3D%22800%22%3EOP%3C%2Ftext%3E%0A%3C%2Fsvg%3E)
Server-side extracted preview paragraphs from the original source.
Developers can now build fast speech-to-speech experiences into their applications
Developers can now build fast speech-to-speech experiences into their applications
Update on August 28, 2025: We announced the general availability of the Realtime API. Learn more here.
Update on February 3, 2025: We no longer limit the number of simultaneous sessions on the Realtime API. Please refer to our docs(opens in a new window) for the latest rate limits on the Realtime API.
Update on October 30, 2024: We've added five new voices with greater range and expressiveness. Cached pricing is now also available for text and audio inputs, lowering the price to $2.50/1M cached text input tokens and $20/1M cached audio input tokens. Learn more here(opens in a new window).
Update on October 17, 2024: Audio inputs and outputs are now available in the Chat Completions API. Get started here(opens in a new window).
Today, we're introducing a public beta of the Realtime API, enabling all paid developers to build low-latency, multimodal experiences in their apps. Similar to ChatGPT’s Advanced Voice Mode, the Realtime API supports natural speech-to-speech conversations using the six preset voices(opens in a new window) already supported in the API.