Share your thoughts, 1 month free Claude Pro on usSee more

Excel Formula Generation on WikiTableQuestions (WikiTQ) 639 solvable problems (test)

84.19Execution Match (EM)

GPT-4o

Updated 21d ago

Evaluation Results

Method	Links
GPT-4o 2025.03		84.19
GPT-4o 2025.03		79.19
GPT-4o 2025.03		78.25
Qwen2.5-coder 14B 2025.03		54.93
Qwen2.5-coder 14B 2025.03		53.52
Qwen2.5 14B 2025.03		51.02
Qwen2.5 14B 2025.03		50.7
Qwen2.5 14B 2025.03		49.92
Qwen2.5 14B 2025.03		49.45
Qwen2.5 14B 2025.03		47.57
Qwen2.5-coder 14B 2025.03		46.95
Qwen2.5-coder 14B 2025.03		46.95
Qwen2.5 14B 2025.03		46.64
Qwen2.5-coder 14B 2025.03		44.76
Qwen2.5-coder 14B 2025.03		40.22
Qwen2.5-coder 3B 2025.03		28.64
CodeLlama-Instruct 13B 2025.03		21.28
Qwen2.5 3B 2025.03		20.81
Qwen2.5-coder 3B 2025.03		18.94
Qwen2.5 3B 2025.03		17.84
CodeLlama-Instruct 13B 2025.03		16.12
Llama2 13B 2025.03		15.49
Qwen2.5-coder 3B 2025.03		15.34
Qwen2.5-coder 3B 2025.03		15.34
Qwen2.5 3B 2025.03		14.87
Qwen2.5 3B 2025.03		14.55
Qwen2.5-coder 3B 2025.03		13.77
Qwen2.5-coder 3B 2025.03		13.62
Qwen2.5 3B 2025.03		13.62
CodeLlama-Instruct 13B 2025.03		13.15
CodeLlama-Instruct 13B 2025.03		12.83
Qwen2.5 3B 2025.03		12.21
CodeLlama-Instruct 13B 2025.03		12.21
CodeLlama-Instruct 13B 2025.03		10.8
CodeLlama-Instruct 7B 2025.03		9.23
Llama2 7B 2025.03		8.61
CodeLlama-Instruct 7B 2025.03		4.85
Llama2 13B 2025.03		4.54
Llama2 7B 2025.03		3.91
Llama2 13B 2025.03		1.72
Llama2 13B 2025.03		1.56
Llama2 7B 2025.03		1.1
CodeLlama-Instruct 7B 2025.03		0.78
Llama2 7B 2025.03		0.63
Llama2 7B 2025.03		0.63
CodeLlama-Instruct 7B 2025.03		0.47
CodeLlama-Instruct 7B 2025.03		0.47
Llama2 13B 2025.03		0.47
CodeLlama-Instruct 7B 2025.03		0.31
Llama2 13B 2025.03		0.31
Llama2 7B 2025.03		0.16