diff --git a/client/data/config.json b/client/data/config.json
index c7b58d8be..6d371613b 100644
--- a/client/data/config.json
+++ b/client/data/config.json
@@ -49,7 +49,9 @@
     {
       "model": "text-embedding-ada-002",
       "name": "Embedding-2",
-      "price": 0
+      "price": 0,
+      "defaultToken": 500,
+      "maxToken": 3000
     }
   ],
   "QAModel": {
diff --git a/client/src/pages/kb/detail/components/Import/Csv.tsx b/client/src/pages/kb/detail/components/Import/Csv.tsx
index ff418cd9c..f0e0eec39 100644
--- a/client/src/pages/kb/detail/components/Import/Csv.tsx
+++ b/client/src/pages/kb/detail/components/Import/Csv.tsx
@@ -17,6 +17,7 @@ const fileExtension = '.csv';
 
 const CsvImport = ({ kbId }: { kbId: string }) => {
   const { kbDetail } = useUserStore();
+  const maxToken = kbDetail.vectorModel?.maxToken || 2000;
 
   const theme = useTheme();
   const router = useRouter();
@@ -39,7 +40,7 @@ const CsvImport = ({ kbId }: { kbId: string }) => {
     mutationFn: async () => {
       const chunks = files.map((file) => file.chunks).flat();
 
-      const filterChunks = chunks.filter((item) => item.q.length < kbDetail.vectorModel.maxToken);
+      const filterChunks = chunks.filter((item) => item.q.length < maxToken);
 
       if (filterChunks.length !== chunks.length) {
         toast({
diff --git a/client/src/pages/kb/detail/components/Import/Manual.tsx b/client/src/pages/kb/detail/components/Import/Manual.tsx
index e91eff379..1cbde1171 100644
--- a/client/src/pages/kb/detail/components/Import/Manual.tsx
+++ b/client/src/pages/kb/detail/components/Import/Manual.tsx
@@ -12,6 +12,7 @@ type ManualFormType = { q: string; a: string };
 
 const ManualImport = ({ kbId }: { kbId: string }) => {
   const { kbDetail } = useUserStore();
+  const maxToken = kbDetail.vectorModel?.maxToken || 2000;
 
   const { register, handleSubmit, reset } = useForm({
     defaultValues: { q: '', a: '' }
@@ -71,8 +72,8 @@ const ManualImport = ({ kbId }: { kbId: string }) => {
         <Box flex={1} mr={[0, 4]} mb={[4, 0]} h={['50%', '100%']} position={'relative'}>
           <Box h={'30px'}>{'匹配的知识点'}</Box>
           <Textarea
-            placeholder={`匹配的知识点。这部分内容会被搜索，请把控内容的质量。最多 ${kbDetail.vectorModel.maxToken} 字。`}
-            maxLength={kbDetail.vectorModel.maxToken}
+            placeholder={`匹配的知识点。这部分内容会被搜索，请把控内容的质量。最多 ${maxToken} 字。`}
+            maxLength={maxToken}
             h={['250px', '500px']}
             {...register(`q`, {
               required: true,
diff --git a/docSite/assets/imgs/model-m3e1.png b/docSite/assets/imgs/model-m3e1.png
new file mode 100644
index 000000000..ccb082330
Binary files /dev/null and b/docSite/assets/imgs/model-m3e1.png differ
diff --git a/docSite/assets/imgs/model-m3e2.png b/docSite/assets/imgs/model-m3e2.png
new file mode 100644
index 000000000..140bf9f69
Binary files /dev/null and b/docSite/assets/imgs/model-m3e2.png differ
diff --git a/docSite/assets/imgs/model-m3e3.png b/docSite/assets/imgs/model-m3e3.png
new file mode 100644
index 000000000..c68691a8e
Binary files /dev/null and b/docSite/assets/imgs/model-m3e3.png differ
diff --git a/docSite/assets/imgs/model-m3e4.png b/docSite/assets/imgs/model-m3e4.png
new file mode 100644
index 000000000..ba0cfcd0a
Binary files /dev/null and b/docSite/assets/imgs/model-m3e4.png differ
diff --git a/docSite/assets/jsconfig.json b/docSite/assets/jsconfig.json
index 12196fbee..8f397dcb4 100644
--- a/docSite/assets/jsconfig.json
+++ b/docSite/assets/jsconfig.json
@@ -3,8 +3,8 @@
   "baseUrl": ".",
   "paths": {
    "*": [
-    "../../../../../Library/Caches/hugo_cache/modules/filecache/modules/pkg/mod/github.com/gohugoio/hugo-mod-jslibs-dist/popperjs/v2@v2.21100.20000/package/dist/cjs/popper.js/*",
-    "../../../../../Library/Caches/hugo_cache/modules/filecache/modules/pkg/mod/github.com/twbs/bootstrap@v5.3.0+incompatible/js/*"
+    "../../../../../.cache/hugo_cache/modules/filecache/modules/pkg/mod/github.com/gohugoio/hugo-mod-jslibs-dist/popperjs/v2@v2.21100.20000/package/dist/cjs/popper.js/*",
+    "../../../../../.cache/hugo_cache/modules/filecache/modules/pkg/mod/github.com/twbs/bootstrap@v5.3.0+incompatible/js/*"
    ]
   }
  }
diff --git a/docSite/content/docs/custom-models/_index.md b/docSite/content/docs/custom-models/_index.md
new file mode 100644
index 000000000..7f4398fb6
--- /dev/null
+++ b/docSite/content/docs/custom-models/_index.md
@@ -0,0 +1,8 @@
+---
+weight: 0
+title: '本地模型使用'
+description: 'FastGPT 对接本地模型'
+icon: 'model_training'
+draft: false
+images: []
+---
diff --git a/docSite/content/docs/installation/reference/chatglm2.md b/docSite/content/docs/custom-models/chatglm2.md
similarity index 100%
rename from docSite/content/docs/installation/reference/chatglm2.md
rename to docSite/content/docs/custom-models/chatglm2.md
diff --git a/docSite/content/docs/custom-models/m3e.md b/docSite/content/docs/custom-models/m3e.md
new file mode 100644
index 000000000..f738cc76e
--- /dev/null
+++ b/docSite/content/docs/custom-models/m3e.md
@@ -0,0 +1,81 @@
+---
+title: '接入 M3E 向量模型'
+description: ' 将 FastGPT 接入私有化模型 M3E'
+icon: 'model_training'
+draft: false
+toc: true
+weight: 100
+---
+
+## 前言
+
+FastGPT 默认使用了 openai 的 embedding 向量模型，如果你想私有部署的话，可以使用 M3E 向量模型进行替换。M3E 向量模型属于小模型，资源使用不高，CPU 也可以运行。下面教程是基于 “睡大觉” 同学提供的一个 CPU 版本的镜像。
+
+## 部署镜像
+
+镜像名: `stawky/m3e-large-api:latest`  
+端口号: 6008
+
+## 接入 OneAPI
+
+添加一个渠道，参数如下：
+
+![](/imgs/model-m3e1.png)
+
+## 测试
+
+curl 例子：
+
+```bash
+curl --location --request POST 'https://domain/v1/embeddings' \
+--header 'Authorization: Bearer sk-key' \
+--header 'Content-Type: application/json' \
+--data-raw '{
+  "model": "m3e",
+  "input": ["laf是什么"]
+}'
+```
+
+Authorization 为 sk-key。model 为刚刚在 OneAPI 填写的自定义模型。
+
+## 接入 FastGPT
+
+修改 config.json 配置文件，在 VectorModels 中加入 M3E 模型：
+
+```json
+"VectorModels": [
+    {
+      "model": "text-embedding-ada-002",
+      "name": "Embedding-2",
+      "price": 0.2,
+      "defaultToken": 500,
+      "maxToken": 3000
+    },
+    {
+      "model": "m3e",
+      "name": "M3E（测试使用）",
+      "price": 0.1,
+      "defaultToken": 500,
+      "maxToken": 1800
+    }
+],
+```
+
+## 测试使用
+
+1. 创建知识库时候选择 M3E 模型。
+
+注意，一旦选择后，知识库将无法修改向量模型。
+
+![](/imgs/model-m3e2.png)
+
+2. 导入数据
+3. 搜索测试
+
+![](/imgs/model-m3e3.png)
+
+4. 应用绑定知识库
+
+注意，应用只能绑定同一个向量模型的知识库，不能跨模型绑定。并且，需要注意调整相似度，不同向量模型的相似度（距离）会有所区别，需要自行测试实验。
+
+![](/imgs/model-m3e4.png)
diff --git a/docSite/content/docs/installation/docker.md b/docSite/content/docs/installation/docker.md
index 9fc7afac9..f9967e2bd 100644
--- a/docSite/content/docs/installation/docker.md
+++ b/docSite/content/docs/installation/docker.md
@@ -162,72 +162,7 @@ docker-compose up -d
 
 ### 如何自定义配置文件？
 
-需要在 `docker-compose.yml` 同级目录创建一个 `config.json` 文件，内容如下：
-
-```json
-{
-  "FeConfig": {
-    "show_emptyChat": true,
-    "show_register": false,
-    "show_appStore": false,
-    "show_userDetail": false,
-    "show_git": true,
-    "systemTitle": "FastGPT",
-    "authorText": "Made by FastGPT Team.",
-    "gitLoginKey": "",
-    "scripts": []
-  },
-  "SystemParams": {
-    "gitLoginSecret": "",
-    "vectorMaxProcess": 15,
-    "qaMaxProcess": 15,
-    "pgIvfflatProbe": 20
-  },
-  "plugins": {},
-  "ChatModels": [
-    {
-      "model": "gpt-3.5-turbo",
-      "name": "GPT35-4k",
-      "contextMaxToken": 4000,
-      "quoteMaxToken": 2000,
-      "maxTemperature": 1.2,
-      "price": 0,
-      "defaultSystem": ""
-    },
-    {
-      "model": "gpt-3.5-turbo-16k",
-      "name": "GPT35-16k",
-      "contextMaxToken": 16000,
-      "quoteMaxToken": 8000,
-      "maxTemperature": 1.2,
-      "price": 0,
-      "defaultSystem": ""
-    },
-    {
-      "model": "gpt-4",
-      "name": "GPT4-8k",
-      "contextMaxToken": 8000,
-      "quoteMaxToken": 4000,
-      "maxTemperature": 1.2,
-      "price": 0,
-      "defaultSystem": ""
-    }
-  ],
-  "QAModel": {
-    "model": "gpt-3.5-turbo-16k",
-    "name": "GPT35-16k",
-    "maxToken": 16000,
-    "price": 0
-  },
-  "VectorModels": [
-    {
-      "model": "text-embedding-ada-002",
-      "name": "Embedding-2",
-      "price": 0
-    }
-  ]
-}
-```
+需要在 `docker-compose.yml` 同级目录创建一个 `config.json` 文件，内容参考: [配置详解](/docs/installation/reference/configuration/)
 
 然后修改 `docker-compose.yml` 中的 `fastgpt` 容器内容，增加挂载选项即可：
 
diff --git a/docSite/content/docs/installation/reference/configuration.md b/docSite/content/docs/installation/reference/configuration.md
index 3f96bea0e..8b3b7689d 100644
--- a/docSite/content/docs/installation/reference/configuration.md
+++ b/docSite/content/docs/installation/reference/configuration.md
@@ -106,7 +106,9 @@ weight: 751
     {
       "model": "text-embedding-ada-002",
       "name": "Embedding-2",
-      "price": 0
+      "price": 0,
+      "defaultToken": 500,
+      "maxToken": 3000
     }
   ]
 }
diff --git a/docSite/content/docs/installation/upgrading/421.md b/docSite/content/docs/installation/upgrading/421.md
new file mode 100644
index 000000000..dd34b338a
--- /dev/null
+++ b/docSite/content/docs/installation/upgrading/421.md
@@ -0,0 +1,24 @@
+---
+title: '升级到 V4.2.1'
+description: 'FastGPT 从旧版本升级到 V4.2.1 操作指南'
+icon: 'upgrade'
+draft: false
+toc: true
+weight: 763
+---
+
+私有部署，如果添加了配置文件，需要在配置文件中修改 `VectorModels` 字段。增加 defaultToken 和 maxToken，分别对应直接分段时的默认 token 数量和该模型支持的 token 上限（通常不建议超过 3000）
+
+```json
+"VectorModels": [
+    {
+      "model": "text-embedding-ada-002",
+      "name": "Embedding-2",
+      "price": 0,
+      "defaultToken": 500,
+      "maxToken": 3000
+    }
+]
+```
+
+改动目的是，我们认为不需要留有选择余地，选择一个最合适的模型去进行任务即可。
diff --git a/files/deploy/fastgpt/config.json b/files/deploy/fastgpt/config.json
index b4d76a58d..91c46dd22 100644
--- a/files/deploy/fastgpt/config.json
+++ b/files/deploy/fastgpt/config.json
@@ -50,7 +50,9 @@
     {
       "model": "text-embedding-ada-002",
       "name": "Embedding-2",
-      "price": 0
+      "price": 0,
+      "defaultToken": 500,
+      "maxToken": 3000
     }
   ],
   "QAModel": {