Tau3-Voice English Telecom: leaderboard
Sierra's τ³-Voice leaderboard (taubench.com), English edition: voice agents (speech-to-speech models, realtime APIs and cascaded speech-to-text/LLM/text-to-speech pipelines) handle τ³-bench customer-service tasks over simulated phone calls, using tools against a domain database, while a simulated customer speaks to them. Standard submissions with the default v1.0 voice user simulator. Telecom domain: technical support in which the caller must also act on their own device. Score: pass^1 (%), the share of tasks whose final database state matches the goal. Separate from the text Tau3-Bench boards and from the tau-Voice paper's Clean/Realistic boards. Higher is better.
Metric: Pass^1 (%). Source: taubench.com. Saturation forecast: Rough model projection: around 2026. 17 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Phonic 1 (High) | 94.74 | |
| 2 | Pine Voice 1.0 | 91.23 | |
| 3 | Pine Voice Preview | 85.96 | |
| 4 | gpt-live-1-diamond-alpha | 84.21 | |
| 5 | Ava Zero | 79.82 | |
| 6 | Grok Voice 1.0 | 73.68 | |
| 7 | Grok Voice 2.0 (High) | 71.93 | |
| 8 | qwen3.5-omni-plus-realtime | 60.53 | |
| 9 | Grok Voice Fast 1.0 | 40.4 | |
| 10 | Gemini 3.1 Flash Live Preview (Thinking, High) | 21.93 |
Interactive version: theaggregate.ai/benchmark?slug=tau3-voice-english-telecom · How It Works · Data refreshed daily, snapshot 2026-11-05.