Grok Voice 3.0 tops Artificial Analysis speech index at 86.9, sub-second latency
rohanpaul_ai · x · 2026-10-15
xAI's Grok Voice 3.0 tops Artificial Analysis's Speech-to-Speech Index at 86.9, beating OpenAI and Google.
- Index blends spoken reasoning, simulated support calls, blind preference votes, and task success; GPT-Live-1 (Astra, Medium) scores 83.2
- Biggest gain on τ-Voice, which tests agentic calls like rebooking flights or disputing charges: 71.6% vs GPT-Live-1 (Astra) at 74.5%, but 40% lower latency (0.9s vs 1.5s) — the only model over 60% support-call success while starting to speak under 1 second
- Big jump from Grok Voice 2.0's 56.5%
Related event: xAI launches Grok Voice 3.0, tops Speech-to-Speech Index(16 posts)→
More from Models
- Scale CEO spotlights Muse Wonder: untrained for 3D yet beats Gemini 4 in physics visualization tests — alexandr_wang · 2026-10-15
- Cohere tops WMT26 translation benchmark: small agentic model scores 84.36, beating Mistral and DeepL — cohere · 2026-10-15
- Anthropic has never launched a Sonnet/Opus-tier model on a Wednesday — ChrisGPT · 2026-10-15
- 8k context beats 480k: OpenAI Decisions API cuts costs 84% at 98.25% F1 — OpenAIDevs · 2026-10-15
- Gemini 4 Argon hit with HTTP 524 timeouts outside Antigravity; Google payment also glitching — cedric_chee · 2026-10-15
- T3 Code blind audit: Gemini 4 Argon scores 3.95 with 51% false claims, Opus 5.5 tops at 8.40 — theo · 2026-10-15