List your company

Household Planning on Household Domains

96.7Success Rate (PC Assembly)DUPLEX Full System
68.93276.14183.3590.559Mar 25, 2026

Evaluation results

MethodLinks
DUPLEX Full SystemMethod Architecture=Full System, Backbone Model=Qwen3-8B + GPT-4o2026.0396.7—9870.768.783.5
DUPLEX Fast SystemMethod Architecture=Fast System, Backbone Model=Qwen3-8B2026.0376.7—7030.72650.9
LLM+PBackbone Model=GPT-4o2026.0376—40020
LLM-as-PlannerBackbone Model=GPT-4o2026.0370—38.70027.2