GPT Realtime
OpenAI's first general-availability realtime model, capable of responding to audio and text inputs in realtime over WebRTC, WebSocket, or SIP connections.
text + image + audio → text + audio · made by OpenAI
$4→$16per Mtok in / out
Microsoft Azure AI
Sold by 1 way
| Seller | Lane | Rate |
|---|---|---|
| Microsoft Azure AI cloud | standard | $4per Mtok in$16per Mtok out$0.000005per image in$0.4per Mtok cached$0.4per Mtok cached, audio$0.5per Mtok cached, image$32per Mtok in, audio$5per Mtok in, image$64per Mtok out, audioraw.githubusercontent.com · read 2026-09-16 |
Measured 1 standing
| Place | Board | Metric | Score |
|---|---|---|---|
| 14thof 16 | τ³-Voice Leaderboard | Pass^1 | 30.4 |
About
GPT Realtime — a text and audio model from OpenAI, sold by one company from $4 in and $16 out per million tokens, placed 14th of 16 on τ³-Voice Leaderboard.
It takes text, images and audio and returns text and audio, with a context window of 32,000 tokens. The catalogue files it under speak. Only Microsoft Azure AI sells it, at $4 in and $16 out per million tokens. It stands 14th of 16 on τ³-Voice Leaderboard.
Every current figure
- Maker
- OpenAI
- Register
- model
- Takes
- text + image + audio
- Returns
- text + audio
- Context
- 32,000 tokens
- Licence
- not read
- Sellers
- 1
- Maker's own price
- not read
- Price
- $4 in and $16 out per million tokens — Microsoft Azure AI
- Boards
- 1
- Best place
- 14th of 16 — τ³-Voice Leaderboard
Known as 4 names
gpt-realtime-1.0gpt-realtimegpt-realtime-2025-08-28gpt rt 0828