比较AI模型4.7和5.5的性能，认为比4.7好但不如5.5。
i wouldn't go as far as to say slop - i would say it's like 2X better than 4.7 but still falls short of 5.5 in my own usage
Claude is still slop: from one of the few people on the world actually pushing these models to do real things (emulators)