OpenAI Publishes Production-Focused Benchmarks for GPT-Live-1 Voice Agents

OpenAIDevs · x · 2026-09-11

OpenAI's developer account published benchmark results for GPT-Live-1 aimed at production voice agents, focusing on task completion, back-and-forth conversation and turn-taking, response latency, and tool use.

The move signals OpenAI is pitching the model on real-world agent metrics rather than generic leaderboard scores; full results are in the linked post.

Related event: OpenAI launches full-duplex speech model GPT-Live-1 on API(26 posts)→

Original post →

More from Models

Models channel →