← τ-bench leaderboards

Telecom · Standard · pass^2

τ-bench · 8 models

RankModelScore
1Qwen3.5-397B-A17B95.76
2Claude Opus 4.586.11
3Gemini 3 Pro84.5
4Gemini 3 Flash83.48
5GPT-5.282.46
6GLM-576.32
7Claude Sonnet 4.575.44
8GPT-5.242.8363