Initially I aimed to test with at least 10 formulas for each model for SAT/UNSAT, but it turned out to be more expensive than I expected, so I tested ~5 formulas for each case/model. First, I used the openrouter API to automate the process, but I experienced response stops in the middle due to long reasoning process, so I reverted to using the chat interface (I don't if this was a problem from the model provider or if it's an openrouter issue). For this reason I don't have standard outputs for each testing, but I linked to the output for each case I mentioned in results.
The hours that led to bloodshed
，推荐阅读同城约会获取更多信息
Мощный удар Израиля по Ирану попал на видео09:41。旺商聊官方下载对此有专业解读
美国总统特朗普周五表示，已指示美国政府机构停止使用Anthropic的技术。随后，五角大楼宣布将Anthropic列为供应链风险企业。。同城约会对此有专业解读
ВсеРоссияМирСобытияПроисшествияМнения