Tau3-Voice Multilingual Intake (Pass^3): leaderboard
Sierra's τ³-Voice leaderboard (taubench.com), Multilingual edition (τ³-Voice 2.0, the site's default view): voice agents handle τ³-bench customer-service calls with a simulated customer in six languages. Standard submissions only; incomplete submission packages are not shown. Intake domain (the 2.0 entity package), submitted as a separate package and not part of the overall. Score: crossed Pass^3 (%), the share of tasks passed under all three of the benchmark's voice conditions; rows without a Pass^3 file are not shown. Higher is better.
Metric: Crossed Pass^3 (%): task passed in all three voice conditions. Source: taubench.com. Saturation forecast: Rough model projection: around 2026. 3 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | gpt-live-1 | 34 | |
| 2 | Grok Voice 2.0 (High) | 24 | |
| 3 | Gemini 3.8 Live (Extended Thinking) | 19.5 |
Interactive version: theaggregate.ai/benchmark?slug=tau3-voice-multilingual-intake-pass-pow-3 · How It Works · Data refreshed daily, snapshot 2026-11-05.