SpaceXAI says its Grok 4.6 model achieves a score of 61 on the “Artificial Analysis Intelligence Index,” according to a post shared on Hacker News.
The sources provided here do not include details about the index methodology, the test set used, or how the score compares with other models. They also do not specify whether the result is newly measured for Grok 4.6 or refers to an existing benchmark figure. As presented, the reporting focuses on the headline score rather than broader performance metrics such as accuracy, safety evaluations, or real-world task results.
With only limited source material available, there is no clear way to reconcile differing angles across outlets beyond the shared claim that Grok 4.6 is associated with a score of 61 on that particular index.
