Usage guide
Setup, environment variables, and Python/JavaScript/Go usage examples.
OpenAIRealtime
OpenAI Realtime speech-to-speech model that handles listening, reasoning and speaking in one model. Choose it for low-latency voice conversations without a separate STT, LLM and TTS. Authenticates with OPENAI_API_KEY from your environment, or an explicitapi_key.
Fields
str
required
Realtime model to use, e.g. “gpt-realtime-2.1”. Required.
Mapping[str, Any] | None
default:"UNSET"
Session settings such as voice, temperature, turn_detection, input_audio_transcription, tool_choice and modalities (“text”, “audio”). Defaults to None (not set).
str | None
default:"UNSET"
Custom OpenAI API base URL, e.g. for a proxy. Defaults to None (not set).