爱思助手下载链接

deepseek-r1: incentivizing reasoning capability in llms via reinforcement learning