GPT-Realtime-2.1
openai/gpt-realtime-2.1
Realtime speech-to-speech model with configurable reasoning, tool use, and robust voice-agent behavior
Capabilities
ReasoningYes
Tool callingYes
Structured outputNo
TemperatureFixed
AttachmentYes
Open weightsNo
Limits
Context128,000 tokens
Max input96,000 tokens
Max output32,000 tokens
Details
Familygpt
Knowledge cutoff2024-09-30
Release date2026-07-06
Last updated2026-07-06
Modalities
Inputtext, audio, image
Outputtext, audio
Community Rating
—
No ratings yet
Sign in to rate
Pricing
Cost per 1M tokens
InputN/A
OutputN/A