GPT-Realtime-2.1
Speech and audio
gpt-realtime-2.1
$4 text in, $24 text out per 1M tokens
GPT-Realtime-2
OpenAI’s first reasoning realtime voice model. For new builds prefer gpt-realtime-2.1.
Model idgpt-realtime-2
Price$4 text in, $24 text out per 1M tokens
Checked29 September 2026
Where it stands
57newer speech and audio models since its release
- NVIDIA Nemotron 3.5 Transcribe (30B-A4B)28 Oct 2026
- Hy-SimulMT-30B-A3B22 Oct 2026
- Eleven Realtime v1 Max / Mini20 Oct 2026
- Voxtral Mini Transcribe 3 (+ Mini Transcribe Realtime 3)20 Oct 2026
- Voxtral Small Transcribe 320 Oct 2026
The 5 newest are listed. All speech and audio models
- Before it from OpenAI
- GPT-Audio-1.5 (and gpt-audio / gpt-audio-mini)23 Feb 2026
- After it from OpenAI
- GPT-Realtime-2.16 Jul 2026
- Weights
- Closed weights; access is hosted only.
How to call it
Where to use GPT-Realtime-2, with its model id.
| Provider | Model id | Endpoint | Docs |
|---|---|---|---|
| OpenAI API | gpt-realtime-2 |
wss://api.openai.com/v1/realtime |
developers.openai.com |
Pricing
| Text input | $4 per 1M tokens |
|---|---|
| Text output | $24 per 1M tokens |
| Audio input | $32 per 1M tokens |
| Audio output | $64 per 1M tokens |
| Cached input | $0.40 per 1M tokens |
| Image input | $5 per 1M tokens |
As published: “per 1M tokens (USD); cached audio/text input $0.40, cached image $0.50”.
Source: developers.openai.com, checked 29 September 2026.
Specs
- Input
- Text, audio and images
- Output
- Text and audio
- Context window
- 128,000 tokens
- Max output
- 32,000 tokens
- Open weights
- No
- Licence
- Proprietary
- Training cutoff
- September 2024
- Released
- 7 May 2026
- Status
- Current
- Type
- Speech and audio
What stands out
What its maker marketed at launch, and what people found later, each with its source.
Reasoning speech-to-speech model
First OpenAI realtime voice model with configurable reasoning effort (press: ‘GPT-5-class’ reasoning); higher effort adds latency and tokens.
128K-token realtime context
Context grew from 32K (gpt-realtime-1.5) to 128K tokens, with 32K max output, for long voice-agent sessions.
Notes
Launched 2026-05-07 with gpt-realtime-translate and gpt-realtime-whisper (changelog). Superseded two months later by gpt-realtime-2.1 (2026-07-06) at identical prices, but still listed and not deprecated. Realtime endpoint only; function calling and prompt caching. Official launch post (openai.com) returned 403 to our fetcher, so benchmark claims were not read directly; secondary sources quote OpenAI: +15.2% Big Bench Audio vs gpt-realtime-1.5 (high effort), +13.8% Audio MultiChallenge instruction following (xhigh); one blog reports 96.6% absolute Big Bench Audio at xhigh (unconfirmed).
Sources: