第三，量化技术带来的不只是压缩。 4-bit 量化常常被理解为「把模型压小 4 倍以节省存储」，但它真正的意义在于减少 4 倍的内存吞吐量。在端侧设备上，瓶颈往往不是存储空间，而是内存带宽，也就是数据从内存搬运到处理器的速度。量化技术让小模型在带宽受限的手机和笔记本上，获得了决定性的速度优势。
GitHub 频繁宕机，OpenAI 决定自己造替代品
，详情可参考电影
В двух аэропортах на юге России ввели ограничения на полеты14:55
Shoppers faced a surprise jump in grocery inflation last month, as experts warned there was worse to come if there was prolonged war in the Middle East and the odds of a UK interest cut fell sharply.
French AA gaming developer and accessory manufacturer Nacon has filed for insolvency after its majority shareholder Bigben failed to make a loan repayment, the company said in a press release. "To date, the company reports available assets do not allow it to meet its liabilities," Nacon wrote. The objective with insolvency, it said, was to allow "continued operation, protect employees and maintain jobs while renegotiating with its creditors."