Do y'all remember the snake model evaluation test?

Just watched a video from three years ago where Matt Berman tested Bard (yes, remember Bard?) and most of the tests were like summarization or logic tests. If it ever came to coding, it was the snake test, or pong, and those from three years ago could barely get them right. Now with Qwen 27B locally or Opus 5.5 via subscription it's just so far away. And I'm wondering with which eyes I will see this post three years from now. Probably robot eyes.