Web browsing comparison on BrowseComp
16.7Task Completion Score (TCS)GPT-5-mini
Evaluation Results
| Method | Links | |
|---|---|---|
| GPT-5-miniAverage Steps (Stp)=17.6, Input Tokens (Tok In)=482.0, Output Tokens (Tok Out)=12.6, Execution Time=291.22026.05 | 16.7 | |
| DeepSeek-V3.2Average Steps (Stp)=19.7, Input Tokens (Tok In)=340.3, Output Tokens (Tok Out)=1.7, Execution Time=137.02026.05 | 12.9 | |
| Gemini-3-flashAverage Steps (Stp)=7.0, Input Tokens (Tok In)=562.1, Output Tokens (Tok Out)=66.9, Execution Time=813.62026.05 | 10.3 | |
| Qwen3-235B-A22BAverage Steps (Stp)=5.5, Input Tokens (Tok In)=268.4, Output Tokens (Tok Out)=0.2, Execution Time=51.32026.05 | 3.8 |