GPT-Realtime-2, GPT-Realtime-Translate and GPT-Realtime-Whisper
GPT-Realtime-2 is OpenAI's first voice model with GPT-5-class reasoning, joined by live translation and streaming transcription models.
Speech-to-speech model with adjustable reasoning effort (minimal to xhigh, default low), tool calls and interruption handling; Translate handles 70+ input languages into 13 output languages in real time; Whisper streams speech-to-text. GPT-Realtime-2.1 followed 2026-07-06.
- Date
- Thursday, 7 May 2026
- Lab
- OpenAI
- Kind
- model
- Access
- closed API
Figures
| Measure | Value | Measured by |
|---|---|---|
| Translate: languages | 70+ input, 13 output streaming speech translation | company |
Sources
- openai.com/index/advancing-voice-intelligence-with-new-models-in-the-api/
- developers.openai.com/api/docs/changelog
- developers.openai.com/api/docs/models/gpt-realtime-2
This record was checked against its sources on 6 October 2026. How we check