Function Calling on BFCL Single-turn
84.2AccuracyEvoPrompt
Evaluation Results
| Method | Links | |
|---|---|---|
| EvoPromptBackbone=GPT-4o2026.03 | 84.2 | |
| EVOTOOLBackbone=GPT-4o2026.03 | 83.9 | |
| OPROBackbone=GPT-4o2026.03 | 82.5 | |
| PromptBreederBackbone=GPT-4o2026.03 | 81.3 | |
| DRAFTBackbone=GPT-4o2026.03 | 77.8 | |
| ANYTOOLBackbone=GPT-4o2026.03 | 77.1 | |
| EVOTOOLBackbone=Qwen3-8B2026.03 | 76.9 | |
| EASYTOOLBackbone=GPT-4o2026.03 | 76.1 | |
| EvoPromptBackbone=Qwen3-8B2026.03 | 75.7 | |
| OPROBackbone=Qwen3-8B2026.03 | 75.2 | |
| ReActBackbone=GPT-4o2026.03 | 74.8 | |
| PromptBreederBackbone=Qwen3-8B2026.03 | 73.9 | |
| EASYTOOLBackbone=Qwen3-8B2026.03 | 73.8 | |
| DRAFTBackbone=Qwen3-8B2026.03 | 72.5 | |
| ANYTOOLBackbone=Qwen3-8B2026.03 | 71.8 | |
| ReActBackbone=Qwen3-8B2026.03 | 70.6 | |
| AdaPlannerBackbone=GPT-4o2026.03 | 69.2 | |
| AdaPlannerBackbone=Qwen3-8B2026.03 | 67.1 | |
| Plan-and-SolveBackbone=GPT-4o2026.03 | 57.2 | |
| Plan-and-SolveBackbone=Qwen3-8B2026.03 | 55.5 | |
| CoTBackbone=GPT-4o2026.03 | 43.2 | |
| CoTBackbone=Qwen3-8B2026.03 | 41.9 |