AI 导读
但所有流量都跑在中国芯片上，实现的硬件效率和每 token 成本可与 Nvidia GPU 相媲美。继昨天 Jalapeño 发布之后，CUDA 护城河再次受到考验。(3/3) https://t.co/hZDrwPqbi1
正文
But ALL traffic was served on Chinese chips, attaining hardware efficiency and per-token cost comparable to Nvidia GPUs. The cuda moat is being tested once again after Jalapeño's announcement yesterday. (3/3) https://t.co/hZDrwPqbi1
来源：@SemiAnalysis_ · x.com