Multimodal model

GPT Realtime 1.5

gpt-realtime-1.5 is OpenAI’s flagship real-time voice model for audio-in, audio-out use cases like voice agents and customer support, with support for text, audio, and image inputs and text and audio outputs.

Type
Multimodal
Weights
Closed