Its pretty nutty indeed. The model still might be good, but the botting is wild. On that note, one of my favorite benchmarks to watch is simple bench and R! d

Its pretty nutty indeed. The model still might be good, but the botting is wild. On that note, one of my favorite benchmarks to watch is simple bench and R! doesn't perform as well on that benchmark as all the other public benchmarks, so it might be telling of something.