List your company

E-commerce Performance Evaluation on In-House Ecom Dataset

96.82Shopping Guide PerformanceGPT5-Thinking
84.13287.42690.7294.014Dec 8, 2025

Evaluation results

MethodLinks
GPT5-Thinkingsize=medium2025.1296.826991.3770.337481.3389.6778.6777.8970.2573.4758.3384.2785.6779.3369.6746.876.29
CompassMax-V3-Thinkingthinking_mode=true2025.1296.1480.6798.4882.3390.33888982.3380.1976.7182.4574.3394.588990.6771.591.6785.79
CompassMax-V3thinking_mode=false2025.1295.2481.6798.728389.6786.3390.3381.6779.5178.583.126793.4688.679071.58985.14
Deepseek R1model_family=R12025.1291.0171.3898.87278.3380.7884.338279.2568.83726090.7282.6781.9471.0579.5579.1
gemini-2.5-proversion=2.5, tier=pro2025.1287.977.3397.3672.677782.6789.338273.8168.981.8261.6789.468190.3368.1793.6780.89
Qwen3-235B-A22Bparameters=235B, architecture=A22B2025.1287.696787.8669.6782.6775.6788.677974.4966.6776.436190.6585.6784.6768.1784.6778.27
GPT-4o2025.1287.266790.657576.3381.33818275.5167.1164.3162.6770.2476.6783.6766.6761.6774.65
GLM4.52025.1284.6271.3394.4873.338078.6786.678071.7765.5574.075884.528380.6765.56976.54