OpenAI Launches gpt-realtime: A Production-Ready Voice Agent with Native SIP & MCP Support
product#agent🏛️ Official|Analyzed: Apr 7, 2026 20:29•
Published: Apr 7, 2026 14:32
•1 min read
•Qiita OpenAIAnalysis
OpenAI has officially elevated the Realtime API to General Availability, introducing the powerful gpt-realtime model designed for production environments. This release is a massive leap forward for developers, featuring native SIP support for phone networks and seamless Remote MCP server integration for tools. With significant boosts in instruction following and audio quality, this update transforms voice agent development.
Key Takeaways
- •The new gpt-realtime model shows a 48% improvement in instruction following and 34% better tool calling accuracy compared to the preview.
- •New high-fidelity voices 'Cedar' and 'Marin' have been added, offering top-tier audio quality for new projects.
- •Developers can now connect to public telephone networks via Twilio SIP and integrate remote tools via URL using the new MCP support.
Reference / Citation
View Original"OpenAI announced in April 2026 that the Realtime API has been upgraded to General Availability (GA), officially releasing the new model gpt-realtime. Compared to the traditional preview version, there are three major changes: SIP phone support, Remote MCP server support, and asynchronous function calling."