List your company

Task-goal completion on AppWorld Challenge Qwen-2.5-32B (test)

51Average Task Completion ScoreADAPO
46.8447.924950.08Mar 12, 2026

Evaluation results

MethodLinks
ADAPOAlgorithm=ADAPO, Backbone=Qwen-2.5-32B2026.035159
LOOPAlgorithm=LOOP, Backbone=Qwen-2.5-32B2026.035054
DAPOAlgorithm=DAPO, Backbone=Qwen-2.5-32B2026.035056
RLOOAlgorithm=RLOO, Backbone=Qwen-2.5-32B2026.034750