它记得你之前说过的It Remembers What You Said Earlier
微型指南 · 约 4 分钟Guide · about 4 min · 引用核验 2026-11-03（仅引用链接与日期） · refs checked 2026-11-03 (reference links and dates only)
它能看到的范围叫上下文窗口，超出部分不在输入里。What it can see is the context window; anything outside is not in its input.
一次对话里，它始终记得前面说过的所有内容。Within one conversation it always remembers everything said earlier.
模型生成时使用本次请求提供的上下文；产品还可能检索历史或保存偏好。是否跨会话保留、保存什么和怎样删除，要看具体产品，不能由一次接话正确推断。Generation uses context supplied for the request; a product may also retrieve history or store preferences. Cross-session retention and deletion depend on the product, not on one correct continuation.
例子与判断Example and interpretation
以下是教学情境，不是本站运行或实测记录。A teaching scenario, not a measured or executed site test.
在新会话里要求沿用上次的表格列名时，先确认产品是否保存了相应记忆，或直接提供列名。一次答对不能证明它保留了所有历史；一次答错也不能证明产品完全没有记忆功能。When asking a new chat to reuse table columns, check whether the product saved them or supply them again. One correct answer does not prove complete history retention; one mistake does not prove the absence of memory features.
- 各成员的上下文长度（千问 3.8）Context length per member (Qwen3.8) ↗
AI 常见误区总索引AI Misconceptions: Index · 怎么向 AI 提问How to Ask AI · 什么时候不该用 RAGWhen Not to Use RAG · 上下文工程怎么做How to Do Context Engineering