Web Shopping on AgentBench Web Shopping
36.2Task Completion Score (TCS)GPT-5-mini
Evaluation Results
| Method | Links | |
|---|---|---|
| GPT-5-miniAverage Steps (Stp)=14.5, Input Tokens (Tok In)=80.1, Output Tokens (Tok Out)=4.6, Execution Time=82.12026.05 | 36.2 | |
| Gemini-3-flashAverage Steps (Stp)=10.6, Input Tokens (Tok In)=76.3, Output Tokens (Tok Out)=2.9, Execution Time=210.72026.05 | 26.6 | |
| Qwen3-235B-A22BAverage Steps (Stp)=5.9, Input Tokens (Tok In)=25.3, Output Tokens (Tok Out)=0.1, Execution Time=75.02026.05 | 10.7 | |
| DeepSeek-V3.2Average Steps (Stp)=19.9, Input Tokens (Tok In)=93.5, Output Tokens (Tok Out)=0.9, Execution Time=149.42026.05 | 4 |