Introducing GPT-Realtime-2 in the API: Our smartest speech model yet, bringing GPT-5 level reasoning capabilities to voice agents. Voice agents are now real-time collaborators, capable of listening, reasoning, and solving complex problems during conversations. This model is now live in the API alongside the streaming models GPT-Realtime-Translate and GPT-Realtime-Whisper — a new set of audio capabilities designed for next-gen voice interaction interfaces.