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

Fact Checking on TableBench (test)

85.42Accuracy

Qwen3-4B

Updated 13d ago

Evaluation Results

Method	Links
Qwen3-4B 2026.05		85.42
Qwen3-8B 2026.05		85.42
Qwen3-8B 2026.05		84.38
Qwen3-4B 2026.05		82.29
Qwen3-8B 2026.05		82.29
Qwen3-8B 2026.05		82.29
Qwen3-8B 2026.05		82.29
Qwen3-4B 2026.05		81.25
Qwen3-4B 2026.05		81.25
Qwen3-4B 2026.05		79.17
Qwen3-8B 2026.05		79.17
Qwen3-8B 2026.05		78.12
Qwen3-4B 2026.05		77.08
Qwen3-4B 2026.05		76.04
Qwen3-4B 2026.05		76.04
Qwen3-8B 2026.05		75
Qwen2.5-Coder-7B-Instruct 2026.05		73.96
Qwen2.5-Coder-7B-Instruct 2026.05		70.83
Qwen2-7B-Instruct 2026.05		69.79
TableLLM-CodeQwen-7B 2026.05		68.75
Qwen2.5-Coder-7B-Instruct 2026.05		67.71
Qwen2.5-Coder-7B-Instruct 2026.05		67.71
TableLLM-CodeQwen-7B 2026.05		67.71
TableLLM-CodeQwen-7B 2026.05		67.71
TableLLM-Llama3.1-8B 2026.05		67.71
TableLLM-Qwen2-7B 2026.05		67.71
TableLLM-CodeQwen-7B 2026.05		66.67
TableLLM-DeepseekCoder-7B 2026.05		66.67
TableLLM-DeepseekCoder-7B 2026.05		65.96
Qwen2.5-Coder-7B-Instruct 2026.05		65.62
TableLLM-Llama3.1-8B 2026.05		65.62
TableLLM-Llama3.1-8B 2026.05		65.62
TableLLM-CodeQwen-7B 2026.05		64.58
TableLLM-DeepseekCoder-7B 2026.05		64.58
TableLLM-Llama3.1-8B 2026.05		64.58
TableLLM-Llama3.1-8B 2026.05		64.58
TableLLM-Qwen2-7B 2026.05		64.58
Qwen2-7B-Instruct 2026.05		63.54
Qwen2.5-Coder-7B-Instruct 2026.05		63.54
TableLLM-DeepseekCoder-7B 2026.05		63.54
TableLLM-Llama3-8B 2026.05		63.54
TableLLM-Llama3-8B 2026.05		63.54
TableLLM-Llama3.1-8B 2026.05		63.54
TableLLM-Qwen2-7B 2026.05		63.54
DeepSeek-R1-Distill-Llama-8B 2026.05		62.5
Meta-Llama-3-8B-Instruct 2026.05		62.5
TableLLM-Llama3-8B 2026.05		62.5
TableLLM-Llama3-8B 2026.05		62.5
TableLLM-Qwen2-7B 2026.05		62.5
Qwen2-7B-Instruct 2026.05		61.46
Qwen3-1.7B 2026.05		61.46
TableLLM-DeepseekCoder-7B 2026.05		61.46
TableLLM-Llama3-8B 2026.05		61.46
TableLLM-Llama3-8B 2026.05		61.46
TableLLM-Llama3.1-8B 2026.05		61.46
TableLLM-Qwen2-7B 2026.05		61.46
TableLLM-Qwen2-7B 2026.05		61.46
Qwen2-7B-Instruct 2026.05		60.42
TableLLM-DeepseekCoder-7B 2026.05		60.42
TableLLM-DeepseekCoder-7B 2026.05		60.42
TableLLM-Llama3-8B 2026.05		60.42
DeepSeek-R1-Distill-Llama-8B 2026.05		59.38
DeepSeek-R1-Distill-Llama-8B 2026.05		59.38
Meta-Llama-3-8B-Instruct 2026.05		59.38
Qwen2.5-Coder-7B-Instruct 2026.05		59.38
Qwen3-1.7B 2026.05		59.38
TableLLM-CodeQwen-7B 2026.05		59.38
TableLLM-DeepseekCoder-7B 2026.05		59.38
TableLLM-Llama3-8B 2026.05		59.38
TableLLM-Qwen2-7B 2026.05		59.38
Qwen2.5-7B-Instruct 2026.05		58.33
Qwen3-1.7B 2026.05		57.29
TableLLM-CodeQwen-7B 2026.05		57.29
Meta-Llama-3-8B-Instruct 2026.05		56.25
Meta-Llama-3-8B-Instruct 2026.05		56.25
Qwen2.5-7B-Instruct 2026.05		56.25
Llama-3.1-8B-Instruct 2026.05		55.21
Meta-Llama-3-8B-Instruct 2026.05		55.21
Qwen2.5-7B-Instruct 2026.05		55.21
Qwen3-1.7B 2026.05		55.21
TableLLM-Llama3.1-8B 2026.05		55.21
DeepSeek-R1-Distill-Llama-8B 2026.05		54.17
Qwen2.5-Coder-7B-Instruct 2026.05		54.17
TableGPT2-7B 2026.05		54.17
TableGPT2-7B 2026.05		53.12
Meta-Llama-3-8B-Instruct 2026.05		52.08
Qwen2.5-7B-Instruct 2026.05		52.08
DeepSeek-R1-Distill-Llama-8B 2026.05		51.04
Llama-3.1-8B-Instruct 2026.05		51.04
Qwen2.5-7B-Instruct 2026.05		51.04
Qwen3-1.7B 2026.05		51.04
Meta-Llama-3-8B-Instruct 2026.05		50
Qwen2.5-7B-Instruct 2026.05		50
Qwen2-7B-Instruct 2026.05		48.96
TableGPT2-7B 2026.05		48.96
Llama-3.1-8B-Instruct 2026.05		47.92
DeepSeek-R1-Distill-Llama-8B 2026.05		46.88
Qwen2-7B-Instruct 2026.05		46.88
Llama-3.1-8B-Instruct 2026.05		45.83
Qwen3-1.7B 2026.05		45.83

Showing 100 of 136 rows