Skip to content

Commit 235b7c9

Browse files
author
antalike
committed
Add .env.intl and update cdn url
1 parent cc240c3 commit 235b7c9

17 files changed

Lines changed: 75 additions & 33 deletions

File tree

‎content/cn/memos_cloud/introduction/algorithm.md‎

Lines changed: 4 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -53,7 +53,8 @@ desc: 让大模型从一次性对话工具，进化为真正具有长期记忆
5353

5454
MemOS 的设计核心，是把「记忆」作为一个独立系统层，和计算、存储一样，成为 AI 应用的基础能力。它的整体架构可以概括为 **三层结构**： <span style="color: rgb(61, 170, 214);">**API 与应用接口层、记忆调度与管理层、记忆存储与基础设施层**</span>
5555

56-
![art.gif](https://statics.memtensor.com.cn/memos/art.gif)
56+
::cdn-img{src="/memos/art.gif" alt="art.gif"}
57+
::
5758

5859
* 在 <span style="color: rgb(61, 170, 214);">**API 与应用接口层**</span>，MemOS 提供了标准化的 Memory API，开发者可以通过简单的接口实现<span style="color: rgb(61, 170, 214);">**记忆创建、删除、更新**</span>等操作，让大模型具备易于调用和扩展的持久记忆能力，支持多轮对话、长期任务和跨会话个性化等复杂应用场景。
5960

@@ -94,7 +95,8 @@ MemOS 的设计核心，是把「记忆」作为一个独立系统层，和计
9495
* 依托对对话轮次与时间差的精准感知，系统能够智能预测下一个场景中可能被调用的记忆内容，并动态路由与预加载所需的明文、参数和激活记忆，从而在生成阶段即刻命中，最大化信息引入的效率和推理的流畅性。
9596

9697

97-
![640.gif](https://statics.memtensor.com.cn/memos/ani.gif)
98+
::cdn-img{src="/memos/ani.gif" alt="640.gif"}
99+
::
98100

99101
<figure style="width: fit-content">
100102
<img src="https://cdn.memtensor.com.cn/img/1758687680524_waiu4s_compressed.png " alt="记忆调度的核心思路" />

‎content/cn/open_source/best_practice/mcp_for_cozespace_and_tools.md‎

Lines changed: 14 additions & 7 deletions
Original file line numberDiff line numberDiff line change
@@ -205,13 +205,15 @@ Connected to MCP server
205205

206206
### 步骤1: 打开Coze空间并进入工具配置页面
207207

208-
![Coze空间配置页面](https://statics.memtensor.com.cn/memos/coze_space_1.png)
208+
::cdn-img{src="memos/coze_space_1.png" alt="Coze空间配置页面"}
209+
::
209210

210211
### 步骤2: 添加自定义MCP工具
211212

212213
在工具配置页面中添加自定义工具：
213214

214-
![添加自定义工具](https://statics.memtensor.com.cn/memos/coze_space_2.png)
215+
::cdn-img{src="memos/coze_space_2.png" alt="添加自定义工具"}
216+
::
215217

216218
### 步骤3: 配置MCP连接地址
217219

@@ -255,15 +257,18 @@ python src/memos/api/server_api.py --port 8001
255257
1. 在Coze中选择"IDE插件"创建方式
256258
2. 配置请求到您部署的Server API服务
257259

258-
![Coze IDE插件配置](https://statics.memtensor.com.cn/memos/coze_tools_1.png)
260+
::cdn-img{src="memos/coze_tools_1.png" alt="Coze IDE插件配置"}
261+
::
259262

260263
### 步骤3: 实现add_memory工具
261264

262-
![配置add_memory操作](https://statics.memtensor.com.cn/memos/coze_tools_2.png)
265+
::cdn-img{src="memos/coze_tools_2.png" alt="配置add_memory操作"}
266+
::
263267

264268
**代码示例：** IDE中配置`add_memory`操作并发布：
265269

266-
![配置add_memory操作](https://statics.memtensor.com.cn/memos/coze_tools_2.png)
270+
::cdn-img{src="memos/coze_space_2.png" alt="配置add_memory操作"}
271+
::
267272
详细代码如下
268273

269274
```python
@@ -330,7 +335,8 @@ def chat_handler(args: Args[Input]) -> Output:
330335

331336
发布完成后，可以在"我的资源"中查看插件：
332337

333-
![发布后的插件资源](https://statics.memtensor.com.cn/memos/coze_tools_3.png)
338+
::cdn-img{src="memos/coze_tools_3.png" alt="发布后的插件资源"}
339+
::
334340

335341
### 步骤5: 集成到智能体工作流
336342

@@ -403,7 +409,8 @@ def chat_handler(args: Args[Input]) -> Output:
403409

404410
发布完成后，可以在"我的资源"中查看插件，以插件形式融入智能体工作流：
405411

406-
![发布后的插件资源](https://statics.memtensor.com.cn/memos/coze_tools_3.png)
412+
::cdn-img{src="memos/coze_tools_3.png" alt="发布后的插件资源"}
413+
::
407414

408415
### 构建智能体并测试
409416

‎content/cn/open_source/contribution/overview.md‎

Lines changed: 2 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -11,5 +11,6 @@ desc: 欢迎阅读 MemOS 贡献指南！了解如何配置开发环境、遵循
1111

1212
你的贡献让这个项目更加优秀！✨ 如果有任何问题，欢迎提交 issue 或参与讨论，也可以扫描下方二维码加入我们的 Discord 或微信社群与我们联系。
1313

14-
<img src="https://statics.memtensor.com.cn/memos/qr-code.png" alt="QR Code" width="600">
14+
::cdn-img{src="/memos/qr-code.png" alt="QR Code" width="600"}
15+
::
1516

‎content/cn/open_source/contribution/writing_docs.md‎

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -21,7 +21,7 @@ desc: 本项目使用 Nuxt Content 构建支持 Markdown 和富 Vue 组件的文
2121
- `banner`（字符串） - 页面顶部展示的横幅图片链接
2222
- `links`（数组） - 包含标签、URL 和图标的相关链接数组
2323

24-
![Frontmatter 示例](https://statics.memtensor.com.cn/memos/frontmatter.png)
24+
::cdn-img{src="/memos/frontmatter.png" alt="Frontmatter 示例"}
2525
::
2626

2727
**完整 Frontmatter 示例：**

‎content/cn/open_source/cookbook/chapter3/overview.md‎

Lines changed: 4 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -6,7 +6,8 @@ title: 使用 MemOS 构建智能小说分析系统
66

77
在开始编码之前，让我们看看MemOS到底解决了什么问题：
88

9-
![Cookbook-Chapter3-Chart](https://statics.memtensor.com.cn/memos/cookbook-chapter3-chart.png)
9+
::cdn-img{src="/memos/cookbook-chapter3-chart.png" alt="Cookbook-Chapter3-Chart"}
10+
::
1011

1112
**实际效果对比举例：**
1213

@@ -42,7 +43,8 @@ MemOS就像是专业的"调料品牌"，它已经为我们准备好了：
4243

4344
**核心技术架构**：
4445

45-
![Cookbook-Chapter3-Core](https://statics.memtensor.com.cn/memos/cookbook-chapter3-core.png)
46+
::cdn-img{src="/memos/cookbook-chapter3-core.png" alt="Cookbook-Chapter3-Core"}
47+
::
4648

4749
**数据处理流水线**：
4850

‎content/cn/open_source/home/memos_intro.md‎

Lines changed: 2 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -3,7 +3,8 @@ title: 什么是 MemOS？
33
desc: "**MemOS** 是为大语言模型 (LLMs) 和智能体打造的**记忆操作系统**。它将记忆视为**可管理、调度和解释的一级资源**，而不是隐藏在模型权重内部的不透明层。"
44
---
55

6-
![MemOS Architecture](https://statics.memtensor.com.cn/memos/memos-architecture.png)
6+
::cdn-img{src="/memos/memos-architecture.png" alt="MemOS Architecture"}
7+
::
78

89

910
随着 LLMs 的发展，它们需要处理复杂任务——如多轮对话、长期规划、决策制定和个性化用户体验——**赋予它们结构化、管理和演进记忆的能力**对于实现真正的长期智能和适应性变得至关重要。

‎content/cn/open_source/home/overview.md‎

Lines changed: 2 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -22,7 +22,8 @@ links:
2222

2323
**MemOS** 通过将记忆重新定义为具有统一结构、生命周期管理和调度策略的核心一级资源来解决这些挑战。它提供了一个 Python 包，为基于 LLM 的应用程序提供统一的记忆层，实现持久化、结构化和高效的记忆操作。这使 LLMs 具备长期知识保留、强大的上下文管理和记忆增强推理能力，支持更智能和自适应的行为。
2424

25-
![MemOS Architecture](https://statics.memtensor.com.cn/memos/memos-architecture.png)
25+
::cdn-img{src="/memos/memos-architecture.png" alt="MemOS Architecture"}
26+
::
2627

2728
## 主要特性
2829

‎content/en/memos_cloud/introduction/algorithm.md‎

Lines changed: 3 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -54,7 +54,8 @@ In short, MemOS aims to:
5454
At its core, MemOS treats “memory” as an independent system layer—like computation and storage—becoming a fundamental capability for AI applications.
5555
Its overall architecture can be summarized as a **three-layer structure**: <span style="color: rgb(61, 170, 214);">**API & Application Interface Layer, Memory Scheduling & Management Layer, Memory Storage & Infrastructure Layer**</span>
5656

57-
![art.gif](https://statics.memtensor.com.cn/memos/art.gif)
57+
::cdn-img{src="/memos/art.gif" alt="art.gif"}
58+
::
5859

5960
* In the <span style="color: rgb(61, 170, 214);">**API & Application Interface Layer**</span>, MemOS provides standardized Memory APIs. Developers can perform operations such as <span style="color: rgb(61, 170, 214);">**memory creation, deletion, and updating**</span> through simple interfaces, giving large models persistent memory capabilities for multi-turn conversations, long-term tasks, and cross-session personalization.
6061

@@ -95,7 +96,7 @@ From Next-Token Prediction to Next-Scene Prediction
9596
* With precise awareness of conversation turns and time gaps, the system intelligently predicts which memory elements may be needed in the next scene. It dynamically routes and preloads the required plaintext, parameter, and activating memories, ensuring immediate hits during generation and maximizing efficiency and fluency of reasoning.
9697

9798

98-
![640.gif](https://statics.memtensor.com.cn/memos/ani.gif)
99+
::cdn-img{src="/memos/ani.gif" alt="640.gif"}
99100

100101
<figure style="width: fit-content">
101102
<img src="https://cdn.memtensor.com.cn/img/1758687680524_waiu4s_compressed.png " alt="Core idea of memory scheduling" />

‎content/en/open_source/best_practice/mcp_for_cozespace_and_tools.md‎

Lines changed: 13 additions & 7 deletions
Original file line numberDiff line numberDiff line change
@@ -205,13 +205,15 @@ After the service is deployed, configure the MCP connection in Coze Space.
205205

206206
### Step 1: Open Coze Space and go to the tool configuration page
207207

208-
![Coze Space configuration page](https://statics.memtensor.com.cn/memos/coze_space_1.png)
208+
::cdn-img{src="/memos/coze_space_1.png" alt="Coze Space configuration page"}
209+
::
209210

210211
### Step 2: Add a custom MCP tool
211212

212213
Add a custom tool on the tool configuration page:
213214

214-
![Add a custom tool](https://statics.memtensor.com.cn/memos/coze_space_2.png)
215+
::cdn-img{src="/memos/coze_space_2.png" alt="Add a custom tool"}
216+
::
215217

216218
### Step 3: Configure the MCP endpoint URL
217219

@@ -257,15 +259,17 @@ python src/memos/api/server_api.py --port 8001
257259
1. In Coze, choose the "IDE plugin" creation method
258260
2. Configure requests to your deployed Server API service
259261

260-
![Coze IDE plugin configuration](https://statics.memtensor.com.cn/memos/coze_tools_1.png)
262+
::cdn-img{src="/memos/coze_tools_1.png" alt="Coze IDE plugin configuration"}
261263

262264
### Step 3: Implement the add_memory tool
263265

264-
![Configure add_memory operation](https://statics.memtensor.com.cn/memos/coze_tools_2.png)
266+
::cdn-img{src="/memos/coze_tools_2.png" alt="Configure add_memory operation"}
267+
::
265268

266269
**Code example:** configure and publish the `add_memory` operation in the IDE:
267270

268-
![Configure add_memory operation](https://statics.memtensor.com.cn/memos/coze_tools_2.png)
271+
::cdn-img{src="/memos/coze_tools_2.png" alt="Configure add_memory operation"}
272+
::
269273

270274
Full code is as follows:
271275

@@ -333,7 +337,8 @@ def chat_handler(args: Args[Input]) -> Output:
333337

334338
After publishing, you can view the plugin under "My Resources":
335339

336-
![Published plugin resource](https://statics.memtensor.com.cn/memos/coze_tools_3.png)
340+
::cdn-img{src="/memos/coze_tools_3.png" alt="Published plugin resource"}
341+
::
337342

338343
### Step 5: Integrate into agent workflow
339344

@@ -406,7 +411,8 @@ No matter which option you choose, MemOS can provide your agent with powerful me
406411

407412
After publishing, you can view the plugin under "My Resources" and integrate it into the agent workflow as a plugin:
408413

409-
![Published plugin resource](https://statics.memtensor.com.cn/memos/coze_tools_3.png)
414+
::cdn-img{src="/memos/coze_tools_3.png" alt="Published plugin resource"}
415+
::
410416

411417
### Build an Agent and Test
412418

‎content/en/open_source/contribution/overview.md‎

Lines changed: 2 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -11,4 +11,5 @@ desc: Welcome to the MemOS contribution guide! Learn how to set up your developm
1111

1212
Your contributions make this project better! ✨ If you have any questions, feel free to open an issue or join the discussion or scan the QR codes below to connect with us on Discord or WeChat.
1313

14-
<img src="https://statics.memtensor.com.cn/memos/qr-code.png" alt="QR Code" width="600">
14+
::cdn-img{src="/memos/qr-code.png" alt="QR Code" width="600"}
15+
::

0 commit comments

Comments
 (0)