From the source
Elon Musk’s artificial intelligence company SpaceXSI, formerly xAI, today released Grok Voice 3.0 , a new version of the speech-to-speech model behind its Voice Agent API.
SpaceXSI is pitching the model to businesses that handle customer support and sales calls with voice agents.
Independent testing firm Artificial Analysis ranks it first among models of its type.
Artificial Analysis, which bases its rankings on reasoning tests, task success and listener votes in blind head-to-head calls, rated Grok Voice 3.0 at 86.9 on its Speech-to-Speech Index .
On the same index, OpenAI Group PBC’s GPT-Live-1, tested with OpenAI’s GPT-6 Astra model handling the reasoning, scored 83.2, and Google LLC’s Gemini 3.8 Live landed at 82.6.
Full Duplex Bench, which grades how naturally a model copes with pauses and interruptions, gave the new Grok model 97%.
The OpenAI model kept a narrow lead on Tau-Voice, a test built around support calls with heavy accents and background noise, at 74.5% against 71.6%.
SpaceXSI says its model got to first audio about 40% faster on that test.
…