List your company

Workflow Generation and Execution on FlowBench 141 cases (694 tests)

0.355CRateES-P&E
0.045080.125540.2060.28646Feb 12, 2026

Evaluation results

MethodLinks
ES-P&EModel=GPT-52026.020.3550.36910.80850.8085
P&EModel=GPT-52026.020.3260.3570.43260.7660.4326
ES-P&EModel=GPT-4.12026.020.2980.3330.96450.92910.9078
ES-ReActModel=GPT-4.12026.020.2620.3040.97870.85820.8369
ReActModel=GPT-52026.020.2620.30.68790.66670.5461
ES-P&EModel=Qwen3-14B2026.020.2410.2750.89360.94330.8511
P&EModel=GPT-4.12026.020.2410.3040.95750.97160.9504
ES-ReActModel=GPT-52026.020.2340.2870.78720.94330.7376
ES-P&EModel=Qwen3-8B2026.020.2130.2590.88650.87230.7872
ES-ReActModel=Qwen3-14B2026.020.2130.2670.74470.9220.7163
ES-ReActModel=Qwen3-32B2026.020.2130.2410.73050.73760.5745
ReActModel=GPT-4.12026.020.2130.2330.74260.84890.7194
P&EModel=Qwen3-14B2026.020.2060.2650.31210.75180.305
ReActModel=Qwen3-32B2026.020.1930.2180.70.40.3857
P&EModel=Qwen3-32B2026.020.1910.2510.26240.46810.2128
ES-P&EModel=Qwen3-32B2026.020.1910.2490.89360.87230.7943
ES-ReActModel=Qwen3-8B2026.020.1840.2320.64540.73050.5177
P&EModel=Qwen3-8B2026.020.1770.210.27660.73050.234
ReActModel=Qwen3-14B2026.020.1490.160.80850.34040.3121
ReActModel=Qwen3-8B2026.020.0570.0620.63830.34040.2553