List your company

Shopping Planning on Shopping Planning

98.3Case AccuracyPIVOT
7.330.92554.5578.175May 11, 2026

Evaluation results

MethodLinks
PIVOTBackbone=Opus 4.6, Feedback regime=w/ HITL2026.0598.3
PIVOTBackbone=Opus 4.5, Feedback regime=w/ HITL2026.0594.2
PIVOTBackbone=Haiku 4.5, Feedback regime=w/ HITL2026.0589.2
PIVOTBackbone=Sonnet 4.5, Feedback regime=w/ HITL2026.0589.2
PIVOTBackbone=Q3 235B, Feedback regime=w/ HITL2026.0576.7
Self-CritiqueBackbone=Opus 4.62026.0570
Self-CritiqueBackbone=Sonnet 4.52026.0555.8
PIVOTBackbone=Opus 4.6, Feedback regime=w/o HITL2026.0555.8
SE-AgentBackbone=Opus 4.62026.0553.7
Self-CritiqueBackbone=Opus 4.52026.0553.3
AgentDebugBackbone=Opus 4.62026.0553.3
ReACTBackbone=Opus 4.62026.0550.8
PIVOTBackbone=Opus 4.5, Feedback regime=w/o HITL2026.0548.3
AgentDebugBackbone=Sonnet 4.52026.0547.5
SE-AgentBackbone=Sonnet 4.52026.0545.8
AgentDebugBackbone=Opus 4.52026.0545.8
SE-AgentBackbone=Opus 4.52026.0543.7
ReACTBackbone=Opus 4.52026.0543.3
PIVOTBackbone=Sonnet 4.5, Feedback regime=w/o HITL2026.0543.3
ReACTBackbone=Sonnet 4.52026.0540.8
PIVOTBackbone=Haiku 4.5, Feedback regime=w/o HITL2026.0537.5
Self-CritiqueBackbone=Haiku 4.52026.0535.8
AgentDebugBackbone=Haiku 4.52026.0533.3
SE-AgentBackbone=Haiku 4.52026.0531.7
ReACTBackbone=Haiku 4.52026.0525.8
PIVOTBackbone=Q3 235B, Feedback regime=w/o HITL2026.0520
Self-CritiqueBackbone=Q3 235B2026.0516.7
SE-AgentBackbone=Q3 235B2026.0515.8
AgentDebugBackbone=Q3 235B2026.0515
ReACTBackbone=Q3 235B2026.0510.8