diff --git a/en/basic/ai/ai-chat.mdx b/en/basic/ai/ai-chat.mdx
index 655acc2..93774c8 100644
--- a/en/basic/ai/ai-chat.mdx
+++ b/en/basic/ai/ai-chat.mdx
@@ -113,6 +113,57 @@ After uploading files, AI can answer questions using both file content and table
- Read this Excel file and identify the fields that should be imported.
- Generate a product description from this image.
+### Memory
+
+Memory stores information you explicitly ask Cuppy to remember. Memory is scoped by user and Base: each user in each Base has an independent memory context, and memory saved in another Base will not automatically appear in the current one.
+
+#### Save Memory in the Current Base
+
+Say what Cuppy should remember in the conversation:
+
+```text
+Please remember xxx.
+```
+
+#### Reuse Memory from Another Base
+
+Ask Cuppy to read another Base and save the relevant parts into the current Base:
+
+```text
+Cuppy, please read the memory for Base ID bsexxxxxxxxxxxx and save the relevant memory into this current Base.
+```
+
+### Handle Long Tasks
+
+AI Chat is designed for foreground, interactive work in the visible conversation. Do not rely on background tasks or silent execution inside the chat for long-running work.
+
+
+
+ If the interface still shows `Thinking`, or progress updates are still being returned, foreground execution is still running. If the task shows `Completed`, the current foreground execution has ended.
+
+
+ After the AI runtime is restarted, only memory files can be read. Background tasks, temporary files, and intermediate results are not preserved.
+
+
+ In the current environment, the conversation environment has idle cleanup and maximum lifetime limits: after foreground execution ends, an idle environment is released after about 30 minutes, and a single conversation environment has an approximate maximum lifetime of 5 hours. Split long-running tasks into smaller batches. Before starting each batch, agree with AI Chat on the visible output for that batch, such as writing results back to Teable, exporting results, or generating a downloadable file. After confirming that output is complete, continue with the next batch.
+
+
+
+For long-running or repeatable work, use the feature that matches the workflow:
+
+
+
+ Use for record-by-record work, such as large table processing, batch classification, content processing, or field filling.
+
+
+ Use for repeatable work after records are created or updated, notifications, or scheduled data processing.
+
+
+
+
+Teable is improving the experience and runtime capability for long-running tasks. At this stage, plan long-running tasks around the limits above.
+
+
## Prompting Tips
- **State the goal first**: In the first message, describe what to do, the relevant conditions, and the expected output. For example, analyze weekly sales trends over the past 3 months, grouped by region, and return a trend table with a short conclusion.
diff --git a/zh/basic/ai/ai-chat.mdx b/zh/basic/ai/ai-chat.mdx
index 953b45a..68bcacd 100644
--- a/zh/basic/ai/ai-chat.mdx
+++ b/zh/basic/ai/ai-chat.mdx
@@ -113,6 +113,57 @@ AI 可以根据已经查询到的数据生成一次性的 HTML 预览，适合
- 读取这个 Excel，帮我找出需要导入的字段。
- 根据这张图片生成产品描述。
+### 记忆
+
+记忆用于保存你明确要求 Cuppy 记住的信息。记忆会按用户和数据库隔离：每个用户在每个数据库中都有独立的记忆，保存在其他数据库中的记忆不会自动出现在当前数据库。
+
+#### 记住当前数据库的信息
+
+在对话中直接说明要记住的内容：
+
+```text
+请记住 xxx。
+```
+
+#### 复用其他数据库的记忆
+
+让 Cuppy 读取指定数据库，并把相关内容保存到当前数据库：
+
+```text
+Cuppy，请读取数据库 ID bsexxxxxxxxxxxx 的记忆，并把相关记忆保存到当前数据库。
+```
+
+### 处理长任务
+
+AI Chat 更适合在当前可见对话中执行前台交互式任务。不要依赖对话中的后台任务或静默运行来处理长时间工作。
+
+
+
+ 如果界面仍显示 `Thinking`，或仍在持续返回进度，说明前台执行仍在进行。如果任务显示 `Completed`，说明本次前台执行已经结束。
+
+
+ AI 运行环境被重启后，只有记忆文件可以被读取；后台任务、临时文件和中间结果都不会被保存。
+
+
+ 在当前环境下，对话环境存在闲置释放和最长生命周期限制：前台执行结束后，闲置环境大约会在 30 分钟后被释放；单次对话环境也有大约 5 小时的最长生命周期。长时间任务应拆成较小批次处理。每开始一批前，建议先和 AI 约定本批的可见产出，例如写回 Teable、导出结果，或生成可下载文件；确认产出完成后，再继续下一批。
+
+
+
+对于长时间或重复性任务，建议根据处理方式选择更合适的功能：
+
+
+
+ 适合逐条记录处理，例如大型表格处理、批量分类、内容处理或填充字段。
+
+
+ 适合在记录创建或更新后重复执行、发送通知，或按计划处理数据。
+
+
+
+
+Teable 正在改进长时间任务的体验和运行能力。在当前阶段，请按以上限制规划长时间任务。
+
+
## 提问建议
- **先说明目标**：第一条消息尽量写清楚你想做什么、有哪些条件、希望输出成什么样，例如分析过去 3 个月每周销售趋势，按地区分组，并输出一张趋势表和简短结论。