Introducing GPT-Realtime in the API: Livestream on August 28

[vb]

The new and improved Realtime model and API are live.
Learn more in the official announcement topic!

Introducing gpt-realtime and Realtime API updates for production voice agents

The Realtime API is officially out of beta and ready for your production voice agents! We’re also introducing gpt-realtime—our most advanced speech-to-speech model yet—plus new voices and API capabilities: Remote MCPs Image input SIP phone calling Reusable prompts gpt-realtime was trained with customers to excel at real-world tasks like support, personal assistance, and education. It’s better at: • Following instruct…

The hot AI summer continues with an update to the successful realtime API! Or is this bigger?

Let’s see what’s been cooking under the radar other than the new Codex IDE extensions!

Let’s watch the live stream together: 2025-08-28T17:00:00Z

Introducing gpt-realtime in the API

Join Brad Lightcap, Peter Bakkum, Beichen Li, Liyu Chen, Julianne Roberson, and Srini Gopalan as they introduce and demo our most advanced speech-to-speech model and new API features like MCP, SIP, image input, and more.

The stream is live. Waiting for the event to start.

This topic will be updated as the event unfolds.

Commenting is usually deactivated on YouTube during the event. Feel free to share your thoughts and impressions here!

OpenAI

[_j]

Somebody wrote three weeks ago…

@_j

What I anticipated that hasn’t been delivered on API: unified output modality.

I suppose it will be a bit more wait for a gpt-5-audio-preview

[vb]

Apparently it’s a complete model, gpt-realtime.
Who knows what’s possible when 4o Realtime gets upgraded to GPT-5 Realtime?

[office.gs]

What is URL for this voice interface?

[vb]

That was a live demo using the playground:

https://platform.openai.com/audio/realtime

[vb]

GPT-5 realtime in the docs:
https://platform.openai.com/docs/models/gpt-realtime

[office.gs]

@office.gs

What is URL for this voice interface?

thanks. My interface does not have the option to attach a file to a dialogue, but this option was demonstrated in the presentation.

[vb]

I see.
My suggestion is to follow-up in the official announcement topic.

Introducing gpt-realtime and Realtime API updates for production voice agents

The Realtime API is officially out of beta and ready for your production voice agents! We’re also introducing gpt-realtime—our most advanced speech-to-speech model yet—plus new voices and API capabilities: Remote MCPs Image input SIP phone calling Reusable prompts gpt-realtime was trained with customers to excel at real-world tasks like support, personal assistance, and education. It’s better at: • Following instruct…