List your company

Sister Tasks on extended tau-Bench Airline domain

74Pass@1 RatePolicyBank
-1.9217.7937.557.21Apr 16, 2026

Evaluation results

MethodLinks
PolicyBankModel=Gemini-3-Pro2026.0474605248
PolicyBankModel=Claude-4.5-Opus2026.0472555340
ReasoningBankModel=Claude-4.5-Opus2026.0445322816
SynapseModel=Claude-4.5-Opus2026.0433151410
AWMModel=Claude-4.5-Opus2026.0431171110
No memoryModel=Claude-4.5-Opus2026.043018139
ReasoningBankModel=Gemini-3-Pro2026.0423171010
SynapseModel=Gemini-3-Pro2026.042100
No memoryModel=Gemini-3-Pro2026.041000
AWMModel=Gemini-3-Pro2026.041000