Data Analysis on DABench
81.32Pass@1ReAct
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| ReActBackbone=DeepSeek-V3.12025.09 | 81.32 | 89.49 | |
| DATAMINDBackbone=Qwen-2.5-Coder-14B2025.09 | 80.29 | 88.72 | |
| ReActBackbone=o4-mini2025.09 | 79.12 | 86.77 | |
| ReActBackbone=DeepSeek-R12025.09 | 78.73 | 87.55 | |
| ReActBackbone=GPT-52025.09 | 78.21 | 85.21 | |
| DATAMINDBackbone=Qwen-2.5-Coder-7B2025.09 | 77.3 | 87.94 | |
| ReActBackbone=GPT-4o2025.09 | 76.39 | 84.44 | |
| ReActBackbone=Qwen-2.5-72B2025.09 | 75.33 | 86.38 | |
| ReActBackbone=Qwen-2.5-Coder-32B2025.09 | 73.15 | 81.32 | |
| ReActBackbone=Qwen-2.5-Coder-14B2025.09 | 71.21 | 83.27 | |
| ReActBackbone=QwQ-32B2025.09 | 70.17 | 85.21 | |
| ReActBackbone=Llama-3.3-70B2025.09 | 69.78 | 80.16 | |
| Table-R1Backbone=Qwen-2.5-Coder-14B2025.09 | 45.33 | 79.38 | |
| Table-R1Backbone=Qwen-2.5-Coder-7B2025.09 | 42.54 | 78.99 | |
| TableLLMBackbone=Qwen-2.5-Coder-14B2025.09 | 38.26 | 74.71 | |
| TableLLMBackbone=Qwen-2.5-Coder-7B2025.09 | 36.71 | 71.98 | |
| SQL-R1Backbone=Qwen-2.5-Coder-14B2025.09 | 27.24 | 40.47 | |
| OmniSQLBackbone=Qwen-2.5-Coder-7B2025.09 | 26.46 | 36.19 | |
| OmniSQLBackbone=Qwen-2.5-Coder-14B2025.09 | 26.46 | 39.3 | |
| SQL-R1Backbone=Qwen-2.5-Coder-7B2025.09 | 24.9 | 34.63 | |
| ReActBackbone=Qwen-2.5-Coder-7B2025.09 | 15.05 | 35.41 |