List your company

Aggregate Data Analysis Performance on DABench, TableBench, and BIRD Combined

71.16Pass@1DATAMIND
8.86425.03741.2157.383Sep 29, 2025

Evaluation results

MethodLinks
DATAMINDBackbone=Qwen-2.5-Coder-14B2025.0971.1680.25
ReActBackbone=DeepSeek-V3.12025.0970.5879.76
ReActBackbone=GPT-52025.0969.4476.26
ReActBackbone=o4-mini2025.0969.0677.93
DATAMINDBackbone=Qwen-2.5-Coder-7B2025.0968.179.07
ReActBackbone=DeepSeek-R12025.0967.8377.75
ReActBackbone=Qwen-2.5-72B2025.0967.0277.36
ReActBackbone=GPT-4o2025.0963.8573.96
ReActBackbone=Llama-3.3-70B2025.0961.4573.03
ReActBackbone=QwQ-32B2025.0959.4274.87
ReActBackbone=Qwen-2.5-Coder-32B2025.0958.4971.25
ReActBackbone=Qwen-2.5-Coder-14B2025.0956.6471.05
OmniSQLBackbone=Qwen-2.5-Coder-14B2025.0942.4153.13
SQL-R1Backbone=Qwen-2.5-Coder-14B2025.0942.1652.7
OmniSQLBackbone=Qwen-2.5-Coder-7B2025.0941.1750.91
SQL-R1Backbone=Qwen-2.5-Coder-7B2025.0940.8350.5
Table-R1Backbone=Qwen-2.5-Coder-7B2025.0936.5352.03
Table-R1Backbone=Qwen-2.5-Coder-14B2025.0935.8450.79
TableLLMBackbone=Qwen-2.5-Coder-14B2025.0935.2359.89
TableLLMBackbone=Qwen-2.5-Coder-7B2025.0929.953.03
ReActBackbone=Qwen-2.5-Coder-7B2025.0911.2627.58