Our new X account is live! Follow @wizwand_team for updates

Structure Comprehending on RealHitBench

82.71Exact Match (EM)

DeepSeek-R1

Updated 4d ago

Evaluation Results

Method
DeepSeek-R1 2025.06	82.71	84.62
QwQ-32B 2025.12	76.08	-
Qwen3-14B 2025.12	73.02	-
Qwen3-32B 2025.12	71.76	-
DeepSeek-V3 2025.12	71.25	-
Qwen3-72B 2025.12	69.47	-
GPT4o(TreeThinker) 2025.06	66.31	77.42
GPT4o 2025.06	65.91	74.49
Doubao-1.5-pro-32k 2025.06	65.9	72.45
GPT4o(TreeThinker) 2025.06	64.4	75.67
Qwen3-8B 2025.12	64.12	-
TableGPT-R1-8B 2025.12	64.12	-
Gemini1.5-pro 2025.06	63.64	69.71
GPT4o 2025.06	63.04	71.14
Qwen-Plus 2025.12	62.85	-
GPT-4o 2025.12	61.83	-
Llama3.2-90B-Vision-Instruct 2025.06	59.8	70.23
Llama3.2-90B-Vision-Instruct 2025.06	58.52	69.57
Gemini1.5-pro 2025.06	56.74	65.8
Llama3.3-70B-Instruct 2025.06	55.81	68.93
Qwen2.5-72B-Instruct 2025.06	54.55	68.34
TableLLM 2025.12	53.28	-
TableLLM-Llama3.1-8B 2025.06	53.28	58.72
GPT4o(TreeThinker) 2025.06	49.21	58.32
Qwen2-VL-7B-Instruct 2025.06	48.6	60.42
TableGPT2-7B 2025.06	48.23	56.68
Qwen2-VL-7B-Instruct 2025.06	45.29	55.99
GPT4o 2025.06	42.68	52.89
Gemini1.5-pro 2025.06	41.52	50.23
mPLUG-Owl3-7B 2025.06	41.24	48.86
TableLlama 2025.06	36.31	42.91
TableLLM-Qwen2-7B 2025.06	36.09	44.75
Llama-3.1-8B 2025.12	35.9	-
Llama3.1-8B-Instruct 2025.06	35.9	50.8
TableGPT2-7B 2025.12	34.86	-
Llama3.2-90B-Vision-Instruct 2025.06	33.33	46.4
Table-R1-Zero-7B 2025.12	28.5	-
Llama3.2-11B-Vision-Instruct 2025.06	27.99	43.85
Qwen2.5-7B-Instruct 2025.06	23.48	44.81
Llama3.2-11B-Vision-Instruct 2025.06	23.39	43.22
Mistral-7B-Instruct-v0.3 2025.06	21.21	49.62
Llama3.2-11B-Vision-Instruct 2025.06	20.87	32.57
DeepSeek-R1-Distill-Qwen-7B 2025.06	19.08	28.1
DeepSeek-R1-Distill-Llama-8B 2025.06	14.76	26.66
LLaVa-v1.5-7B 2025.06	13.23	30.51
QwQ-32B 2025.06	12.99	37.96
DeepSeek-R1-Distill-Qwen-1.5B 2025.06	9.92	19.93
mPLUG-Owl2-7B 2025.06	9.14	14.21
Table-LLava-7B 2025.06	7.38	11.83