Ha! So we were not looking at the same chart. That makes more sense. > Anthropic did post an official explanation, stating the original chart used a "simpler m

Ha! So we were not looking at the same chart. That makes more sense. > Anthropic did post an official explanation, stating the original chart used a "simpler methodology" that "underestimated Sonnet 5's performance." The new chart supposedly uses their "standard methodology." Oops!