List your company

Stateful Agent-User Interaction on tau-Bench Airline

39.1Pass@1EVOTOOL
6.96415.30723.6531.993Mar 5, 2026

Evaluation results

MethodLinks
EVOTOOLBackbone=GPT-4o2026.0339.1
AdaPlannerBackbone=GPT-4o2026.0338.8
OPROBackbone=GPT-4o2026.0336.8
Plan-and-SolveBackbone=GPT-4o2026.0336.7
EvoPromptBackbone=GPT-4o2026.0336.3
ANYTOOLBackbone=GPT-4o2026.0336.2
ReActBackbone=GPT-4o2026.0335.9
PromptBreederBackbone=GPT-4o2026.0331.3
EASYTOOLBackbone=GPT-4o2026.0329.9
DRAFTBackbone=GPT-4o2026.0329.2
CoTBackbone=GPT-4o2026.0325.3
AdaPlannerBackbone=Qwen3-8B2026.0316.4
EVOTOOLBackbone=Qwen3-8B2026.0315.7
Plan-and-SolveBackbone=Qwen3-8B2026.0315
OPROBackbone=Qwen3-8B2026.0315
EvoPromptBackbone=Qwen3-8B2026.0314.7
ANYTOOLBackbone=Qwen3-8B2026.0314.6
ReActBackbone=Qwen3-8B2026.0314.4
PromptBreederBackbone=Qwen3-8B2026.0311.3
EASYTOOLBackbone=Qwen3-8B2026.0310.3
DRAFTBackbone=Qwen3-8B2026.038.9
CoTBackbone=Qwen3-8B2026.038.2