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

Table Question Answering on ReasonTabQA 1.0 (test)

67.58Accuracy

Gemini-3-Pro-Preview

Updated 1mo ago

Evaluation Results

Method	Links
Gemini-3-Pro-Preview 2026.01		67.58
Claude-Opus-4.5 2026.01		66.2
Doubao-1.5-thinking-pro 2026.01		64.48
Qwen3-8B-Think-SFT-TabCodeRL 2026.01		61.89
Qwen3-32B-Instruct 2026.01		58.76
GPT-4o 2026.01		56.9
Deepseek-R1 2026.01		55.8
OpenAI o1-mini 2026.01		55.8
QWQ-32B 2026.01		54.1
Qwen2.5-72B-Instruct 2026.01		51.92
Qwen3-8B-Instruct 2026.01		49.87
TableGPT2-7B 2026.01		42.05
Qwen3-8B-Instruct 2026.01		40.97
Llama-3.1-8B-Instruct 2026.01		33.24
TableLLM-7B 2026.01		13.5
TableLlama-7B 2026.01		13.23
Deepseek-R1-Distill-Qwen-7B 2026.01		11.86