Tau-Bench Telecom — leaderboard
Agentic customer service benchmark in a simulated telecom environment. Models handle complex multi-turn conversations with tool-calling. From Sierra AI (tau2-bench).
Metric: Pass@1 (%). Source: taubench.com. Status: saturated. 12 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Qwen 3 Max (Thinking) | 98.2 |
| 2 | Claude Sonnet 4.5 | 98 |
| 3 | Gemini 3 Pro | 98 |
| 4 | DeepSeek V3.2 | 96.2 |
| 5 | GPT-5 | 95.8 |
| 6 | Qwen 3 Max | 84.2 |
| 7 | Kimi K2 | 65.8 |
| 8 | O4 Mini | 50.2 |
| 9 | Claude 3.7 Sonnet | 49 |
| 10 | GPT-4.1 Mini | 48.9 |
| 11 | GPT-4.1 | 34 |
Interactive version: theaggregate.ai/benchmark?slug=tau-bench-telecom · How the rankings work · Data refreshed daily, snapshot 2026-07-22.