GPT-Realtime-2.1 Mini
GPT-Realtime-2.1 Mini is a distilled voice-reasoning model for lower-cost interactive audio. It supports connections over WebRTC, WebSocket and SIP.
A smaller voice model retains a large session window
The published window is 128,000 tokens with up to 32,000 output tokens. The model supports tools and prompt caching. Its audio rates are $10 per million input tokens and $20 per million output tokens; text is billed at different rates.GPT-Realtime-2.1 Mini official model documentation
Use a clearly bounded voice task
Practical guidance. For appointment requests, define the details the assistant must collect and the point at which a person must confirm them. Compare task completion, correction handling and turn latency. Do not judge a voice model only by the naturalness of its first reply.
Specifications & access
| Detail | Value and conditions |
|---|---|
| Access | API |
| Model identifiers | gpt-realtime-2.1-mini |
| Context window | 128,000 tokens Published model context; input and output share this budget GPT-Realtime-2.1 Mini official model documentation |
| Maximum output | 32,000 tokens Published maximum output; bounded by the remaining context GPT-Realtime-2.1 Mini official model documentation |
| Input and output | text, audio, image in; text, audio outGPT-Realtime-2.1 Mini official model documentation |
| API routes | Realtime (v1/realtime)GPT-Realtime-2.1 Mini official model documentation |
Related models
- GPT-Realtime-2.1
OpenAI · GPT Realtime
GPT-Realtime-2.1 combines conversational audio with reasoning and tool use.
Launch date not confirmedAPI - GPT-Realtime-1.5
OpenAI · GPT Realtime
GPT-Realtime-1.5 supports interactive voice agents with text, audio and image input.
Launch date not confirmedAPI - GPT-Realtime-2
OpenAI · GPT Realtime
GPT-Realtime-2 is a speech-to-speech model with configurable reasoning and tools.
Launch date not confirmedAPI - GPT-Realtime-Translate
OpenAI · GPT Realtime
GPT-Realtime-Translate converts incoming speech into translated speech and text while the source audio is still arriving.
Launch date not confirmedAPI