← τ-bench leaderboards

Airline · Standard · pass^3

τ-bench · 8 models

RankModelScore
1GPT-5.275
2Claude Opus 4.573.5
3GLM-572
3Gemini 3 Flash72
5Qwen3.5-397B-A17B71.5
6Gemini 3 Pro70
7Claude Sonnet 4.554.5
8GPT-5.227