Show HN: ZetaCrush – An Intelligent LLM Leaderboard (zetacrush.com)

ZetaCrush has launched an innovative LLM Leaderboard that evaluates language models based on a proprietary intelligence aptitude test rather than subjective human judgments. This approach aims to provide a more objective assessment of model performance, addressing potential biases commonly found in traditional head-to-head comparisons. The leaderboard currently ranks Gemini 3 Pro at the top with a score of 0.2000, followed by Claude Opus 4.5 and Deepseek, among others. Notably, the specific testing criteria will remain closed-source, ensuring fairness in evaluation. This initiative is significant for the AI/ML community as it introduces a new benchmark for assessing advanced AI capabilities, encompassing not only text processing but also search and vision tasks. The leaderboard highlights the current limitations of leading models, as all are scoring zero on certain advanced capabilities, indicating areas for potential growth and development. By providing an alternative measuring system, ZetaCrush encourages transparency and further innovation in the field of AI, potentially shifting how developers and researchers evaluate and enhance their models.