OpenAI launches GPT-Live outbound phone calls API at $0.05 per minute
OpenAI has made its GPT-Live outbound telephony API generally available, allowing developers to have voice agents dial real phone numbers. The system works through Session Initiation Protocol (SIP) trunk providers, with Telnyx, Twilio, and Ozonetel named as launch partners. These providers handle the connection to the public switched telephone network, while OpenAI manages the audio processing…
Key points
- OpenAI launched GPT-Live outbound telephony API on November 4, 2026, for general availability.
- The API costs $0.05 per minute and works via SIP providers like Telnyx, Twilio, and Ozonetel.
- New features include HD Voice and bidirectional DTMF signaling to navigate phone menus.
The update includes HD Voice support for higher-fidelity audio and full bidirectional DTMF signaling. This allows agents to both send and receive keypad tones, enabling them to navigate interactive voice response (IVR) systems and phone menus. The feature complements existing WebRTC and WebSocket transports for GPT-Live.
The API is priced at $0.05 per minute, a rate that has been in place since GPT-Live-1 reached the API on September 10, 2026. Businesses must also account for carrier charges from their SIP provider, meaning the total cost of a call is the sum of OpenAI’s fee and the provider’s charges. OpenAI is not acting as a phone carrier but is integrating its intelligence layer with established telephony infrastructure.
The story so far
3 episodes →- OpenAI launches GPT-Live outbound phone calls API at $0.05 per minutethis story
OpenAI launches GPT-Live outbound phone calls API at $0.05 per minute
cryptobriefing.com · 4 November 2026
Loading the full article…
This text was published by cryptobriefing.com and written by Diego Almada Lopez. It is reproduced here with attribution so you can read it in full; the rights remain with the publisher. Read it at the source ↗
The headline, key points and digest above were generated by Digest AI's editorial model from the linked sources. Automated summaries can contain errors: the sources are the record. Spotted a mistake? Tell us. Published by Martin K., who runs Digest AI and handles corrections.
More in Agents & Tools
All →- Researchers introduce ReScopeBench to test long-lived AI agents’ experience retention · 1 src
- Researchers propose Agent Harness Awareness to improve agent decision-making with real-time telemetry · 1 src
- The Story of Turning the Process of Creating AI Manga with Claude Code into a Skill (2-Panel Manga) · 1 src
- Developer reports persistent errors and hallucinations in Cursor Composer 2.5 · 1 src
- Google makes Gemini Gems read‑only on personal accounts starting November 17 · 1 src
Comments
via GitHub Discussions