AI News Feed
Market watch
Intero Biosystems Sells Lab-Grown Mini Intestines for Drug Testing as FDA Weighs Animal AlternativesTechRadar Posts October 15 Hints and Answers for Quordle, Strands and ConnectionsApple Maps Launches Detailed City Experience in Austin, Adds F1 Circuit GuideFish Audio Releases Drama 3, a 35B Open-Weight MoE Text-to-Speech ModelSpaceXSI's Grok Voice 3.0 Tops Speech-to-Speech Index, Outscoring OpenAI and GoogleHack of Refrigeration Controllers at 700 French Supermarkets Causes Spoilage and Fires
Intero Biosystems Sells Lab-Grown Mini Intestines for Drug Testing as FDA Weighs Animal AlternativesTechRadar Posts October 15 Hints and Answers for Quordle, Strands and ConnectionsApple Maps Launches Detailed City Experience in Austin, Adds F1 Circuit GuideFish Audio Releases Drama 3, a 35B Open-Weight MoE Text-to-Speech ModelSpaceXSI's Grok Voice 3.0 Tops Speech-to-Speech Index, Outscoring OpenAI and GoogleHack of Refrigeration Controllers at 700 French Supermarkets Causes Spoilage and Fires
Products & Applications

SpaceXSI's Grok Voice 3.0 Tops Speech-to-Speech Index, Outscoring OpenAI and Google

SpaceXSI launches Grok Voice 3.0, ranked first on a speech-to-speech index ahead of OpenAI and Google.

AI News Feed · 2026-10-15 13:01

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. OpenAI’s GPT-Live-1, tested with OpenAI’s GPT-6 Astra model handling reasoning, scored 83.2, while Google’s Gemini 3.8 Live scored 82.6. Full Duplex Bench, which grades how naturally a model handles pauses and interruptions, gave the 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 said its model reached first audio about 40% faster on that test. A task in Artificial Analysis’ Big Bench Audio run cost about 2.5 cents to complete on Grok Voice 3.0, compared with 3 cents on GPT-Live-1 and 5.5 cents on OpenAI’s GPT-Realtime-2.1.

Outside Artificial Analysis’ testing, Grok Voice 3.0 also led for accuracy on ServiceNow’s open-source EVA-Bench, scoring 74% across more than 200 enterprise scenarios. No model did better when accuracy and customer experience were scored together. Nitin Aggarwal, a senior director of product management at ServiceNow, said the model “pairs top accuracy with a natural, responsive conversation.”

In production, Tesla vehicles already account for millions of Grok Voice conversations a day, with drivers using it to check calendars and answer email. Larger tasks such as booking dinner go through Grok Bot, and its voice mode runs on the new model as well. At Starlink, SpaceX’s satellite internet business, Grok Voice answers about 20,000 customer calls a day in several languages. SpaceXSI said roughly 70% are resolved without a human stepping in, and one in four sales calls ends in a purchase before the caller hangs up.

SpaceX acquired xAI in February. The AI unit was renamed SpaceXAI in July and took on the SpaceXSI name earlier this month. Grok Voice 3.0 is available now in the Voice Agent API at 8 cents a minute, or $4.80 an hour, the same price as Grok Voice 2.0. Developers whose agents point to the grok-voice-latest model alias have already been moved to the new version without changes on their end.