AA Speech Agent Arena - Task Success Rate: leaderboard
Artificial Analysis Speech Agent Arena: share of arena conversations in which the voice agent completed the user's task (successes over successes plus failures, overall slice). Task success (%), higher is better.
Metric: Task success (%) in arena conversations. Source: artificialanalysis.ai. Saturation forecast: Rough model projection: around 2026. 30 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Grok Voice 3.0 | 94.83 | |
| 2 | Grok Voice 2.0 (High) | 94.61 | |
| 3 | Gemini 3.8 Live | 93.15 | |
| 4 | Phonic 1 (High) | 92.31 | |
| 5 | Cartesia Managed Agents | 91.74 | |
| 6 | GPT Realtime 2.1 (High) | 91.55 | |
| 7 | GPT-Live-1 (Sol, low) | 90.94 | |
| 8 | ElevenLabs Agents (Default Cascaded System) | 90.54 | |
| 9 | GPT-Realtime-2 (High) | 89.82 | |
| 10 | GPT Realtime (Aug '25) | 89.43 |
Interactive version: theaggregate.ai/benchmark?slug=aa-speech-agent-arena-task-success-rate · How It Works · Data refreshed daily, snapshot 2026-11-05.