List your company

Sister Tasks on extended tau-Bench Retail domain

83Pass@1 RatePolicyBank
24.7639.885570.12Apr 16, 2026

Evaluation results

MethodLinks
PolicyBankModel=Gemini-3-Pro2026.0483726855
PolicyBankModel=Claude-4.5-Opus2026.0478696767
ReasoningBankModel=Gemini-3-Pro2026.0464393324
No memoryModel=Claude-4.5-Opus2026.0447292522
ReasoningBankModel=Claude-4.5-Opus2026.0445363028
AWMModel=Gemini-3-Pro2026.0439221511
SynapseModel=Gemini-3-Pro2026.0436201611
No memoryModel=Gemini-3-Pro2026.04311460
AWMModel=Claude-4.5-Opus2026.0430221610
SynapseModel=Claude-4.5-Opus2026.0427181614