Workflow execution on 7,680 workflows Financial synthetic benchmark benchmark leaderboard · SOTA2 Research

Workflow execution on 7,680 workflows Financial synthetic benchmark

81.5QualityInKH

Evaluation Results

Method	Links
InKH2026.06		81.5	900.2	1,540.3	32.9	0.9	99.9	0.0086
Khnoinv2026.06		76.5	960	1,550.2	32.9	27.1	92.8	0.0087
WikiWalk2026.06		70.7	5,281.1	8,697.3	16.3	27.1	53.8	0.0301
SimpleMem2026.06		65.7	1,931.1	3,039.7	24.6	23.3	50.1	0.0131
ToolAgent2026.06		54	1,351.6	1,279.5	0	0	32	0.0078
ModelOnly2026.06		44	654.5	880	0	0	8	0.0026