OpenAI
GPT-Realtime-2.1
Realtime speech-to-speech model with configurable reasoning, tool use, and robust voice-agent behavior
gpt-realtime-2.1Context
128K
Max output
32K
Input $/M
$4.00
Output $/M
$24.00
Cache read $/M
$0.40
Cache write $/M
—
Against the catalogue
Context window128K
Larger than 25% of the 5,751 models listed
Input price$4.00
Cheaper than 10% of priced models
Capabilities
Reasoningsupported
Tool callingsupported
Structured outputnot supported
Attachmentssupported
Temperaturenot supported
Open weightsnot supported
Interleaved thinkingnot supported
Modalities
Input
TextAudioImage
Output
TextAudio
Reasoning controls
effortminimal, low, medium, high, xhighPricing detail
| Rate | Base $/M |
|---|---|
| Input | $4.00 |
| Output | $24.00 |
| Cache read | $0.40 |
| Audio input | $32.00 |
| Audio output | $64.00 |