Long-term Planning on AgentBench LTP
32.3Task Completion Score (TCS)Gemini-3-flash
Evaluation Results
| Method | Links | |
|---|---|---|
| Gemini-3-flashAverage Steps (Stp)=17.4, Input Tokens (Tok In)=1539.6, Output Tokens (Tok Out)=295.1, Execution Time=1974.02026.05 | 32.3 | |
| GPT-5-miniAverage Steps (Stp)=19.5, Input Tokens (Tok In)=45.4, Output Tokens (Tok Out)=21.7, Execution Time=497.72026.05 | 27.3 | |
| Qwen3-235B-A22BAverage Steps (Stp)=12.0, Input Tokens (Tok In)=26.8, Output Tokens (Tok Out)=0.5, Execution Time=63.82026.05 | 21.6 | |
| DeepSeek-V3.2Average Steps (Stp)=20.4, Input Tokens (Tok In)=51.8, Output Tokens (Tok Out)=1.0, Execution Time=74.82026.05 | 12.6 |