List your company
ResearchFormerly WizWand

Program of Thought Reasoning on TableBench

51.96Rge ScoreGPT-4o
-2.078411.950825.9840.0092Dec 23, 2025
Updated 8mo ago

Evaluation results

MethodLinks
GPT-4oModel Scale Group=Larger2025.1251.96
Qwen-PlusModel Scale Group=Larger2025.1241.79
QwQ-32BModel Scale Group=Larger2025.1240.03
TableGPT2-7BModel Scale Group=Comparable2025.1239.8
Qwen3-32BModel Scale Group=Larger2025.1237.78
Qwen3-14BModel Scale Group=Larger2025.1236.61
TableGPT-R1-8BModel Scale Group=Comparable2025.1235.12
DeepSeek-V3Model Scale Group=Larger2025.1233.05
Qwen3-8BModel Scale Group=Comparable2025.1228.01
Qwen3-72BModel Scale Group=Larger2025.1227.72
Table-R1-Zero-7BModel Scale Group=Comparable2025.127.54
Llama-3.1-8BModel Scale Group=Comparable2025.126.73
TableLLMModel Scale Group=Comparable2025.120