https://www.hansking.cn/blog A posts created with Next.js and Tailwind.css en-us admin@hansking.cn (HANS) admin@hansking.cn (HANS) Sun, 31 Mar 2024 10:00:00 GMT https://www.hansking.cn/blog/2026/ContextWindow-de-nei-cun-mo-xing-yu-cheng-ben-ji-suan https://www.hansking.cn/blog/2026/ContextWindow-de-nei-cun-mo-xing-yu-cheng-ben-ji-suan [{"value":"Context Window 的内存模型与成本计算","url":"#context-window-的内存模型与成本计算","depth":1},{"value":"一、Context Window 包含哪些东西","url":"#一context-window-包含哪些东西","depth":2},{"value":"二、Token 不是字符，语言和内容会改变比例","url":"#二token-不是字符语言和内容会改变比例","depth":2},{"value":"三、为什么上下文会占显存","url":"#三为什么上下文会占显存","depth":2},{"value":"四、并发让内存问题变成乘法","url":"#四并发让内存问题变成乘法","depth":2},{"value":"五、长上下文不等于模型有效利用全文","url":"#五长上下文不等于模型有效利用全文","depth":2},{"value":"六、什么时候应该用检索而不是全文输入","url":"#六什么时候应该用检索而不是全文输入","depth":2},{"value":"七、对话历史应该如何增长","url":"#七对话历史应该如何增长","depth":2},{"value":"八、成本计算要包含输入和输出","url":"#八成本计算要包含输入和输出","depth":2},{"value":"九、服务端要做上下文预算器","url":"#九服务端要做上下文预算器","depth":2},{"value":"十、长上下文压测应该怎么做","url":"#十长上下文压测应该怎么做","depth":2},{"value":"十一、上下文窗口会影响系统架构","url":"#十一上下文窗口会影响系统架构","depth":2},{"value":"十二、我的总结：窗口是预算，不是宣传数字","url":"#十二我的总结窗口是预算不是宣传数字","depth":2}] Sun, 31 Mar 2024 10:00:00 GMT admin@hansking.cn (HANS) AILLMContext WindowKV Cache成本优化 https://www.hansking.cn/blog/2026/KVCacheLLM-tui-li-wei-shen-me-hui-bei-xian-cun-ka-zhu https://www.hansking.cn/blog/2026/KVCacheLLM-tui-li-wei-shen-me-hui-bei-xian-cun-ka-zhu [{"value":"KV Cache：LLM 推理为什么会被显存卡住","url":"#kv-cachellm-推理为什么会被显存卡住","depth":1},{"value":"一、先理解自回归生成","url":"#一先理解自回归生成","depth":2},{"value":"二、KV Cache 为什么会占很多显存","url":"#二kv-cache-为什么会占很多显存","depth":2},{"value":"三、Prefill 和 Decode 对 Cache 的使用不同","url":"#三prefill-和-decode-对-cache-的使用不同","depth":2},{"value":"四、为什么连续批处理需要更聪明的 Cache","url":"#四为什么连续批处理需要更聪明的-cache","depth":2},{"value":"五、前缀缓存可以复用，但必须确认边界","url":"#五前缀缓存可以复用但必须确认边界","depth":2},{"value":"六、Cache 量化和压缩要用质量换算","url":"#六cache-量化和压缩要用质量换算","depth":2},{"value":"七、Cache 管理要考虑取消、抢占和恢复","url":"#七cache-管理要考虑取消抢占和恢复","depth":2},{"value":"八、上下文窗口不是无限记忆","url":"#八上下文窗口不是无限记忆","depth":2},{"value":"九、如何压测 KV Cache","url":"#九如何压测-kv-cache","depth":2},{"value":"总结：缓存是速度和空间的交换","url":"#总结缓存是速度和空间的交换","depth":2}] Sun, 28 Sep 2025 10:00:00 GMT admin@hansking.cn (HANS) AILLMKV CacheGPU推理优化