Grok Voice 3.0 launches and tops the speech-to-speech leaderboard
What happened
xAI released Grok Voice 3.0, a speech-to-speech model, on October 14 and opened it the same day through the Voice Agent API, with pricing unchanged and latency under one second. It ranks first on Artificial Analysis's Speech-to-Speech Index at 86.9 and scores 71.6% on τ-Voice, close to GPT-Live-1 (Astra) with 40% lower latency. The leaderboard puts GPT-Live-1 (Astra, medium) at 83.2 and Gemini 3.8 Live Extended Thinking (High) at 82.6. Grok Voice 3.0 is 5.6 points above the previous generation and tops the list on debut.
Written by AI from the coverage · updated 2 hours ago
Developments
- Oct 15 02:38 · 1 reportxAI releases Grok Voice 3.0, keeps Voice Agent API pricing unchangedAI HOT · Products
- Oct 15 01:26 · 1 reportArtificial Analysis 评测：Grok Voice 3.0 以 86.9 分登顶语音到语音模型榜AI HOT · Models
Coverage
Follow the reports to see the story from different sides.
- AI HOT · ProductsPickxAI releases Grok Voice 3.0, keeps Voice Agent API pricing unchanged
xAI released Grok Voice 3.0, a speech-to-speech model available today through the Voice Agent API with sub-second latency. It ranks first on Artificial Analysis's Speech-to-Speech Index at 86.9 and scores 71.6% on τ-Voice, close to GPT-Live-1 (Astra) but with 40% lower latency.
- AI HOT · ModelsArtificial Analysis 评测：Grok Voice 3.0 以 86.9 分登顶语音到语音模型榜
Grok Voice 3.0 在 Artificial Analysis Speech to Speech Index 以 86.9 分首秀登顶，领先 GPT-Live-1（Astra, medium）的 83.2 和 Gemini 3.8 Live Extended Thinking（High）的 82.6。相较 Grok Voice 2.0，其得分提升 5.6 分。
Heat over time
Not enough continuous observations to draw a trend yet.