AA Speech to Speech - Tau-Voice 2.0 - Telecom: leaderboard

Sierra tau-bench v2.0.0 voice: 50 English tasks each in airline, retail and telecom; three trials; GPT5.5 low simulated caller speaking ElevenLabs eleven_v3; common prompt/tools/policy, 2400-second call limit. State/action/output checks; infrastructure errors excluded and rerun. Overall is equal domain mean. GPT-Live's source-declared backend prompt differs. Separate from V1; no cross-edition score replacement. Metric: Pass@1 (%), mean of three trials. Source: https://artificialanalysis.ai/speech-to-speech

Metric: Pass@1 (%), mean of three trials. Source: artificialanalysis.ai. Saturation forecast: Rough model projection: around 2026. 11 models tracked. Updated 17 Nov 2026.

Top models

#ModelSettingScoreOverall rank
1Phonic 1High95.33
2GPT-Live 1 (Astra, medium)86
3Gemini 3.8 Live (Extended Thinking)High82.67
4Grok Voice 3.075.33
5Eleven Realtime v1 Max70.67
6Eleven Realtime v1 Mini70
7GPT-Live 1 (Sol, low)68
8Qwen Audio 3.0 Realtime Plus59.33
9Gemini 3.1 Flash Live High24.67
10GPT Realtime 2.1High16
11Gemini 3.8 Live14.67

Interactive version: theaggregate.ai/benchmark?slug=aa-speech-to-speech-tau-voice-2-0-telecom · How It Works · Data refreshed daily, snapshot 2026-11-17.