OpenAI Launches GPT-Realtime-2 with GPT-5-Class Reasoning for Voice AI Agents

Summary

OpenAI released three new streaming audio models for its Realtime API: GPT-Realtime-2, GPT-Realtime-Translate, and GPT-Realtime-Whisper. GPT-Realtime-2 is the flagship, bringing GPT-5-class reasoning to a native speech-to-speech model with a 128K token context window — up from 32K — and features including parallel tool calling with audible transparency, conversational preambles, and adjustable reasoning effort across five levels. Independent benchmarks showed instruction retention rising from 36