Grok Voice 3.0 tops OpenAI and Google in benchmarks

SpaceXSI, formerly xAI, has released Grok Voice 3.0, a speech-to-speech model for business voice agents. Independent firm Artificial Analysis ranked it first with a score of 86.9 on its Speech-to-Speech Index, ahead of OpenAI's GPT-Live-1 at 83.2 and Google's Gemini 3.8 Live at 82.6. It also led on EVA-Bench enterprise accuracy at 74% and costs 2.5 cents per task versus 3 cents for GPT-Live-1. At Starlink, it handles 20,000 daily calls with 70% resolved without human help. Pricing remains eight cents per minute.