Grok Voice tops Speech-to-Speech Index, outperforming all GPT Realtime models
XFreeze · x · 2026-08-25
Grok Voice Think Fast 2.0 claimed the #1 spot on Artificial Analysis’ Speech-to-Speech Index, outperforming every GPT Realtime model. The benchmark combines speech reasoning, agentic performance, human preference, and actual task success.
More from Models
- Users complain GPT-5.6 Sol over-engineers tasks with verbose code — Lenox_Shawn · 2026-08-25
- GLM-5.3 shows strong cost-efficiency vs GPT 5.6 on TerminalBench-3.0 — zainhas · 2026-08-25
- Local LLM Test: Qwen2.5 72B Hits 100 tok/s on RTX 4090 — julianharris · 2026-08-25
- Redditor finds MiniMax H3 performs far better with Mandarin prompts than English — apoke890 · 2026-08-25
- Suspected Gemini 3.8 Flash leak surfaces on Reddit — Last_Conclusion_8984 · 2026-08-25
- Train Only Projector to Add New Modalities Without Regressing LLM Capabilities — rohanpaul_ai · 2026-08-25