Sequential Tool Use on RestBench Spotify (Success Rate)
86.1Success RateEVOTOOL
Evaluation Results
| Method | Links | |
|---|---|---|
| EVOTOOLBackbone=GPT-4o2026.03 | 86.1 | |
| DRAFTBackbone=GPT-4o2026.03 | 85.2 | |
| EASYTOOLBackbone=GPT-4o2026.03 | 83.5 | |
| ANYTOOLBackbone=GPT-4o2026.03 | 78.4 | |
| EvoPromptBackbone=GPT-4o2026.03 | 77.4 | |
| OPROBackbone=GPT-4o2026.03 | 76.4 | |
| PromptBreederBackbone=GPT-4o2026.03 | 75.2 | |
| ReActBackbone=GPT-4o2026.03 | 74.3 | |
| EVOTOOLBackbone=Qwen3-8B2026.03 | 73.2 | |
| DRAFTBackbone=Qwen3-8B2026.03 | 72.4 | |
| AdaPlannerBackbone=GPT-4o2026.03 | 70.1 | |
| Plan-and-SolveBackbone=GPT-4o2026.03 | 69.2 | |
| EASYTOOLBackbone=Qwen3-8B2026.03 | 68 | |
| ANYTOOLBackbone=Qwen3-8B2026.03 | 66.6 | |
| EvoPromptBackbone=Qwen3-8B2026.03 | 65.8 | |
| OPROBackbone=Qwen3-8B2026.03 | 64.9 | |
| PromptBreederBackbone=Qwen3-8B2026.03 | 63.9 | |
| CoTBackbone=GPT-4o2026.03 | 63.5 | |
| ReActBackbone=Qwen3-8B2026.03 | 63.2 | |
| AdaPlannerBackbone=Qwen3-8B2026.03 | 59.6 | |
| Plan-and-SolveBackbone=Qwen3-8B2026.03 | 58.8 | |
| CoTBackbone=Qwen3-8B2026.03 | 54 |