← τ-bench leaderboards

Airline · Standard · pass^1

τ-bench · 8 models

RankModelScore
1Claude Opus 4.584
2GPT-5.283
3GLM-582.5
3Gemini 3 Flash82.5
5Qwen3.5-397B-A17B81.5
6Gemini 3 Pro80.5
7Claude Sonnet 4.572
8GPT-5.252.5