List your company

Table Reasoning on TableBench

72.52Pass@1ReAct
9.267225.688642.1158.5314Sep 29, 2025

Evaluation results

MethodLinks
ReActBackbone=DeepSeek-V3.12025.0972.5281.68
ReActBackbone=o4-mini2025.0971.0380.15
DATAMINDBackbone=Qwen-2.5-Coder-14B2025.0970.9581.81
ReActBackbone=GPT-52025.0969.9378.37
ReActBackbone=DeepSeek-R12025.0968.9679.52
DATAMINDBackbone=Qwen-2.5-Coder-7B2025.0967.679.39
ReActBackbone=Qwen-2.5-72B2025.0965.4476.21
ReActBackbone=GPT-4o2025.0964.9775.06
ReActBackbone=Qwen-2.5-Coder-32B2025.0961.1172.26
ReActBackbone=QwQ-32B2025.0957.7975.19
ReActBackbone=Qwen-2.5-Coder-14B2025.0956.9669.97
Table-R1Backbone=Qwen-2.5-Coder-7B2025.0956.3663.61
ReActBackbone=Llama-3.3-70B2025.0955.4770.36
Table-R1Backbone=Qwen-2.5-Coder-14B2025.0950.3858.91
TableLLMBackbone=Qwen-2.5-Coder-14B2025.0946.4476.08
OmniSQLBackbone=Qwen-2.5-Coder-14B2025.0941.9852.67
SQL-R1Backbone=Qwen-2.5-Coder-14B2025.0941.2251.02
TableLLMBackbone=Qwen-2.5-Coder-7B2025.0941.0170.36
SQL-R1Backbone=Qwen-2.5-Coder-7B2025.0940.8450.64
OmniSQLBackbone=Qwen-2.5-Coder-7B2025.0939.9550.25
ReActBackbone=Qwen-2.5-Coder-7B2025.0911.728.63