List your company

Original Tasks on tau-Bench Airline domain extended

70Pass Rate @ Threshold 1ReasoningBank
61.6863.846668.16Apr 16, 2026

Evaluation results

MethodLinks
ReasoningBankModel=Gemini-3-Pro2026.0470636060
PolicyBankModel=Gemini-3-Pro2026.0470646259
ReasoningBankModel=Claude-4.5-Opus2026.0469615348
PolicyBankModel=Claude-4.5-Opus2026.0469595149
SynapseModel=Gemini-3-Pro2026.0467605547
AWMModel=Gemini-3-Pro2026.0467635955
AWMModel=Claude-4.5-Opus2026.0467545148
No memoryModel=Gemini-3-Pro2026.0466625958
No memoryModel=Claude-4.5-Opus2026.0462524744
SynapseModel=Claude-4.5-Opus2026.0462545047