LLMTabBench (Zero-Shot) - Classic Datasets: leaderboard

Metric: ROC-AUC (0-1; no demonstrations; 24 classic OpenML and Kaggle binary classification datasets; generation mode, prompt with column names and task description; mean over datasets). Source: arxiv.org. Saturation forecast: Estimated already saturated. 4 models tracked.

Top models

#ModelScore
1Qwen 3 14B0.73
2GPT-4o Mini0.73
3Qwen 3 8B0.71
4Qwen 3 1.7B0.64

Interactive version: theaggregate.ai/benchmark?slug=llmtabbench-zero-shot-classic-datasets · How It Works · Data refreshed daily, snapshot 2026-09-26.