OpenAI launches GPT-4o for real-time text, vision and voice interaction

Summary

GPT-4o introduced a single end-to-end model for text, image and audio rather than a three-stage speech pipeline. OpenAI reported near-conversational audio latency and lower API cost, while rolling out text and vision first and introducing real-time voice and video capabilities in stages. Capability, benchmark and availability claims are attributed to OpenAI unless independently verified. Follow the original source for complete technical details, current availability, licensing and later updates.