diff --git a/apps/dataset/urls.py b/apps/dataset/urls.py
index 2068922ee..405101796 100644
--- a/apps/dataset/urls.py
+++ b/apps/dataset/urls.py
@@ -23,6 +23,7 @@
     path('dataset/<str:dataset_id>/document/_bach', views.Document.Batch.as_view()),
     path('dataset/<str:dataset_id>/document/batch_hit_handling', views.Document.BatchEditHitHandling.as_view()),
     path('dataset/<str:dataset_id>/document/<int:current_page>/<int:page_size>', views.Document.Page.as_view()),
+    path('dataset/<str:dataset_id>/document/batch_refresh', views.Document.BatchRefresh.as_view()),
     path('dataset/<str:dataset_id>/document/<str:document_id>', views.Document.Operate.as_view(),
          name="document_operate"),
     path('dataset/document/split', views.Document.Split.as_view(),
@@ -34,7 +35,6 @@
          name="document_export"),
     path('dataset/<str:dataset_id>/document/<str:document_id>/sync', views.Document.SyncWeb.as_view()),
     path('dataset/<str:dataset_id>/document/<str:document_id>/refresh', views.Document.Refresh.as_view()),
-    path('dataset/<str:dataset_id>/document/batch_refresh', views.Document.BatchRefresh.as_view()),
     path('dataset/<str:dataset_id>/document/<str:document_id>/paragraph', views.Paragraph.as_view()),
     path(
         'dataset/<str:dataset_id>/document/<str:document_id>/paragraph/migrate/dataset/<str:target_dataset_id>/document/<str:target_document_id>',
diff --git a/apps/dataset/views/document.py b/apps/dataset/views/document.py
index c2ef152a0..d41535b0b 100644
--- a/apps/dataset/views/document.py
+++ b/apps/dataset/views/document.py
@@ -239,7 +239,7 @@ def put(self, request: Request, dataset_id: str, document_id: str):
     class BatchRefresh(APIView):
         authentication_classes = [TokenAuth]
 
-        @action(methods=['POST'], detail=False)
+        @action(methods=['PUT'], detail=False)
         @swagger_auto_schema(operation_summary="批量刷新文档向量库",
                              operation_id="批量刷新文档向量库",
                              request_body=
diff --git a/apps/setting/models_provider/impl/aws_bedrock_model_provider/credential/llm.py b/apps/setting/models_provider/impl/aws_bedrock_model_provider/credential/llm.py
index 2392d3d04..e64d8b282 100644
--- a/apps/setting/models_provider/impl/aws_bedrock_model_provider/credential/llm.py
+++ b/apps/setting/models_provider/impl/aws_bedrock_model_provider/credential/llm.py
@@ -19,7 +19,7 @@ class BedrockLLMModelParams(BaseForm):
                                     precision=2)
 
     max_tokens = forms.SliderField(
-        TooltipLabel('输出最大Tokens', '较高的数值会使输出更加随机，而较低的数值会使其更加集中和确定'),
+        TooltipLabel('输出最大Tokens', '指定模型可生成的最大token个数'),
         required=True, default_value=1024,
         _min=1,
         _max=4096,
diff --git a/apps/setting/models_provider/impl/azure_model_provider/credential/llm.py b/apps/setting/models_provider/impl/azure_model_provider/credential/llm.py
index 1906d20c5..b9e730aa0 100644
--- a/apps/setting/models_provider/impl/azure_model_provider/credential/llm.py
+++ b/apps/setting/models_provider/impl/azure_model_provider/credential/llm.py
@@ -25,7 +25,7 @@ class AzureLLMModelParams(BaseForm):
                                     precision=2)
 
     max_tokens = forms.SliderField(
-        TooltipLabel('输出最大Tokens', '较高的数值会使输出更加随机，而较低的数值会使其更加集中和确定'),
+        TooltipLabel('输出最大Tokens', '指定模型可生成的最大token个数'),
         required=True, default_value=800,
         _min=1,
         _max=4096,
diff --git a/apps/setting/models_provider/impl/deepseek_model_provider/credential/llm.py b/apps/setting/models_provider/impl/deepseek_model_provider/credential/llm.py
index 72f101c42..ee2279bbc 100644
--- a/apps/setting/models_provider/impl/deepseek_model_provider/credential/llm.py
+++ b/apps/setting/models_provider/impl/deepseek_model_provider/credential/llm.py
@@ -25,7 +25,7 @@ class DeepSeekLLMModelParams(BaseForm):
                                     precision=2)
 
     max_tokens = forms.SliderField(
-        TooltipLabel('输出最大Tokens', '较高的数值会使输出更加随机，而较低的数值会使其更加集中和确定'),
+        TooltipLabel('输出最大Tokens', '指定模型可生成的最大token个数'),
         required=True, default_value=800,
         _min=1,
         _max=4096,
diff --git a/apps/setting/models_provider/impl/gemini_model_provider/credential/llm.py b/apps/setting/models_provider/impl/gemini_model_provider/credential/llm.py
index 77e01df39..2612205d4 100644
--- a/apps/setting/models_provider/impl/gemini_model_provider/credential/llm.py
+++ b/apps/setting/models_provider/impl/gemini_model_provider/credential/llm.py
@@ -25,7 +25,7 @@ class GeminiLLMModelParams(BaseForm):
                                     precision=2)
 
     max_tokens = forms.SliderField(
-        TooltipLabel('输出最大Tokens', '较高的数值会使输出更加随机，而较低的数值会使其更加集中和确定'),
+        TooltipLabel('输出最大Tokens', '指定模型可生成的最大token个数'),
         required=True, default_value=800,
         _min=1,
         _max=4096,
diff --git a/apps/setting/models_provider/impl/kimi_model_provider/credential/llm.py b/apps/setting/models_provider/impl/kimi_model_provider/credential/llm.py
index ef2aeb122..1ee2fcee1 100644
--- a/apps/setting/models_provider/impl/kimi_model_provider/credential/llm.py
+++ b/apps/setting/models_provider/impl/kimi_model_provider/credential/llm.py
@@ -25,7 +25,7 @@ class KimiLLMModelParams(BaseForm):
                                     precision=2)
 
     max_tokens = forms.SliderField(
-        TooltipLabel('输出最大Tokens', '较高的数值会使输出更加随机，而较低的数值会使其更加集中和确定'),
+        TooltipLabel('输出最大Tokens', '指定模型可生成的最大token个数'),
         required=True, default_value=1024,
         _min=1,
         _max=4096,
diff --git a/apps/setting/models_provider/impl/ollama_model_provider/credential/llm.py b/apps/setting/models_provider/impl/ollama_model_provider/credential/llm.py
index 14634b478..5558bcab4 100644
--- a/apps/setting/models_provider/impl/ollama_model_provider/credential/llm.py
+++ b/apps/setting/models_provider/impl/ollama_model_provider/credential/llm.py
@@ -23,7 +23,7 @@ class OllamaLLMModelParams(BaseForm):
                                     precision=2)
 
     max_tokens = forms.SliderField(
-        TooltipLabel('输出最大Tokens', '较高的数值会使输出更加随机，而较低的数值会使其更加集中和确定'),
+        TooltipLabel('输出最大Tokens', '指定模型可生成的最大token个数'),
         required=True, default_value=1024,
         _min=1,
         _max=4096,
diff --git a/apps/setting/models_provider/impl/openai_model_provider/credential/llm.py b/apps/setting/models_provider/impl/openai_model_provider/credential/llm.py
index f7d244a53..58dfc1308 100644
--- a/apps/setting/models_provider/impl/openai_model_provider/credential/llm.py
+++ b/apps/setting/models_provider/impl/openai_model_provider/credential/llm.py
@@ -25,7 +25,7 @@ class OpenAILLMModelParams(BaseForm):
                                     precision=2)
 
     max_tokens = forms.SliderField(
-        TooltipLabel('输出最大Tokens', '较高的数值会使输出更加随机，而较低的数值会使其更加集中和确定'),
+        TooltipLabel('输出最大Tokens', '指定模型可生成的最大token个数'),
         required=True, default_value=800,
         _min=1,
         _max=4096,
diff --git a/apps/setting/models_provider/impl/qwen_model_provider/credential/llm.py b/apps/setting/models_provider/impl/qwen_model_provider/credential/llm.py
index 7ad068454..a8177c545 100644
--- a/apps/setting/models_provider/impl/qwen_model_provider/credential/llm.py
+++ b/apps/setting/models_provider/impl/qwen_model_provider/credential/llm.py
@@ -25,7 +25,7 @@ class QwenModelParams(BaseForm):
                                     precision=2)
 
     max_tokens = forms.SliderField(
-        TooltipLabel('输出最大Tokens', '较高的数值会使输出更加随机，而较低的数值会使其更加集中和确定'),
+        TooltipLabel('输出最大Tokens', '指定模型可生成的最大token个数'),
         required=True, default_value=800,
         _min=1,
         _max=2048,
diff --git a/apps/setting/models_provider/impl/vllm_model_provider/credential/llm.py b/apps/setting/models_provider/impl/vllm_model_provider/credential/llm.py
index 7ea3cabe4..97c6217c3 100644
--- a/apps/setting/models_provider/impl/vllm_model_provider/credential/llm.py
+++ b/apps/setting/models_provider/impl/vllm_model_provider/credential/llm.py
@@ -19,7 +19,7 @@ class VLLMModelParams(BaseForm):
                                     precision=2)
 
     max_tokens = forms.SliderField(
-        TooltipLabel('输出最大Tokens', '较高的数值会使输出更加随机，而较低的数值会使其更加集中和确定'),
+        TooltipLabel('输出最大Tokens', '指定模型可生成的最大token个数'),
         required=True, default_value=800,
         _min=1,
         _max=4096,
diff --git a/apps/setting/models_provider/impl/volcanic_engine_model_provider/credential/llm.py b/apps/setting/models_provider/impl/volcanic_engine_model_provider/credential/llm.py
index f918b437d..15fffec2c 100644
--- a/apps/setting/models_provider/impl/volcanic_engine_model_provider/credential/llm.py
+++ b/apps/setting/models_provider/impl/volcanic_engine_model_provider/credential/llm.py
@@ -25,7 +25,7 @@ class VolcanicEngineLLMModelParams(BaseForm):
                                     precision=2)
 
     max_tokens = forms.SliderField(
-        TooltipLabel('输出最大Tokens', '较高的数值会使输出更加随机，而较低的数值会使其更加集中和确定'),
+        TooltipLabel('输出最大Tokens', '指定模型可生成的最大token个数'),
         required=True, default_value=1024,
         _min=1,
         _max=4096,
diff --git a/apps/setting/models_provider/impl/wenxin_model_provider/credential/llm.py b/apps/setting/models_provider/impl/wenxin_model_provider/credential/llm.py
index be294e81a..a77a6303f 100644
--- a/apps/setting/models_provider/impl/wenxin_model_provider/credential/llm.py
+++ b/apps/setting/models_provider/impl/wenxin_model_provider/credential/llm.py
@@ -25,7 +25,7 @@ class WenxinLLMModelParams(BaseForm):
                                     precision=2)
 
     max_tokens = forms.SliderField(
-        TooltipLabel('输出最大Tokens', '较高的数值会使输出更加随机，而较低的数值会使其更加集中和确定'),
+        TooltipLabel('输出最大Tokens', '指定模型可生成的最大token个数'),
         required=True, default_value=1024,
         _min=2,
         _max=2048,
diff --git a/apps/setting/models_provider/impl/xf_model_provider/credential/llm.py b/apps/setting/models_provider/impl/xf_model_provider/credential/llm.py
index 770aff27d..0a6d9a0ac 100644
--- a/apps/setting/models_provider/impl/xf_model_provider/credential/llm.py
+++ b/apps/setting/models_provider/impl/xf_model_provider/credential/llm.py
@@ -25,7 +25,7 @@ class XunFeiLLMModelGeneralParams(BaseForm):
                                     precision=2)
 
     max_tokens = forms.SliderField(
-        TooltipLabel('输出最大Tokens', '较高的数值会使输出更加随机，而较低的数值会使其更加集中和确定'),
+        TooltipLabel('输出最大Tokens', '指定模型可生成的最大token个数'),
         required=True, default_value=4096,
         _min=1,
         _max=4096,
@@ -42,7 +42,7 @@ class XunFeiLLMModelProParams(BaseForm):
                                     precision=2)
 
     max_tokens = forms.SliderField(
-        TooltipLabel('输出最大Tokens', '较高的数值会使输出更加随机，而较低的数值会使其更加集中和确定'),
+        TooltipLabel('输出最大Tokens', '指定模型可生成的最大token个数'),
         required=True, default_value=4096,
         _min=1,
         _max=8192,
diff --git a/apps/setting/models_provider/impl/xinference_model_provider/credential/llm.py b/apps/setting/models_provider/impl/xinference_model_provider/credential/llm.py
index bb17e5c22..6317ff663 100644
--- a/apps/setting/models_provider/impl/xinference_model_provider/credential/llm.py
+++ b/apps/setting/models_provider/impl/xinference_model_provider/credential/llm.py
@@ -19,7 +19,7 @@ class XinferenceLLMModelParams(BaseForm):
                                     precision=2)
 
     max_tokens = forms.SliderField(
-        TooltipLabel('输出最大Tokens', '较高的数值会使输出更加随机，而较低的数值会使其更加集中和确定'),
+        TooltipLabel('输出最大Tokens', '指定模型可生成的最大token个数'),
         required=True, default_value=800,
         _min=1,
         _max=4096,
diff --git a/apps/setting/models_provider/impl/zhipu_model_provider/credential/llm.py b/apps/setting/models_provider/impl/zhipu_model_provider/credential/llm.py
index aee7441f1..dc1d1f191 100644
--- a/apps/setting/models_provider/impl/zhipu_model_provider/credential/llm.py
+++ b/apps/setting/models_provider/impl/zhipu_model_provider/credential/llm.py
@@ -25,7 +25,7 @@ class ZhiPuLLMModelParams(BaseForm):
                                     precision=2)
 
     max_tokens = forms.SliderField(
-        TooltipLabel('输出最大Tokens', '较高的数值会使输出更加随机，而较低的数值会使其更加集中和确定'),
+        TooltipLabel('输出最大Tokens', '指定模型可生成的最大token个数'),
         required=True, default_value=1024,
         _min=1,
         _max=4096,
diff --git a/ui/src/api/document.ts b/ui/src/api/document.ts
index 0653f2d40..d4d9b505b 100644
--- a/ui/src/api/document.ts
+++ b/ui/src/api/document.ts
@@ -3,6 +3,7 @@ import { get, post, del, put, exportExcel } from '@/request/index'
 import type { Ref } from 'vue'
 import type { KeyValue } from '@/api/type/common'
 import type { pageRequest } from '@/api/type/common'
+
 const prefix = '/dataset'
 
 /**
@@ -26,14 +27,14 @@ const listSplitPattern: (
 
 /**
  * 文档分页列表
- * @param 参数  dataset_id,   
+ * @param 参数  dataset_id,
  * page {
-              "current_page": "string",
-              "page_size": "string",
-            }
-* param {
-          "name": "string",
-        }
+ "current_page": "string",
+ "page_size": "string",
+ }
+ * param {
+ "name": "string",
+ }
  */
 
 const getDocument: (
@@ -58,22 +59,22 @@ const getAllDocument: (dataset_id: string, loading?: Ref<boolean>) => Promise<Re
 
 /**
  * 创建批量文档
- * @param 参数 
+ * @param 参数
  * {
-  "name": "string",
-  "paragraphs": [
-    {
-      "content": "string",
-      "title": "string",
-      "problem_list": [
-          {
-            "id": "string",
-              "content": "string"
-          }
-      ]
-    }
-  ]
-}
+ "name": "string",
+ "paragraphs": [
+ {
+ "content": "string",
+ "title": "string",
+ "problem_list": [
+ {
+ "id": "string",
+ "content": "string"
+ }
+ ]
+ }
+ ]
+ }
  */
 const postDocument: (
   dataset_id: string,
@@ -85,13 +86,13 @@ const postDocument: (
 
 /**
  * 修改文档
- * @param 参数 
- * dataset_id, document_id, 
+ * @param 参数
+ * dataset_id, document_id,
  * {
-      "name": "string",
-      "is_active": true,
-      "meta": {}
-    }
+ "name": "string",
+ "is_active": true,
+ "meta": {}
+ }
  */
 const putDocument: (
   dataset_id: string,
@@ -124,6 +125,19 @@ const delMulDocument: (
 ) => Promise<Result<boolean>> = (dataset_id, data, loading) => {
   return del(`${prefix}/${dataset_id}/document/_bach`, undefined, { id_list: data }, loading)
 }
+
+const batchRefresh: (
+  dataset_id: string,
+  data: any,
+  loading?: Ref<boolean>
+) => Promise<Result<boolean>> = (dataset_id, data, loading) => {
+  return put(
+    `${prefix}/${dataset_id}/document/batch_refresh`,
+    { id_list: data },
+    undefined,
+    loading
+  )
+}
 /**
  * 文档详情
  * @param 参数 dataset_id
@@ -180,14 +194,14 @@ const delMulSyncDocument: (
 
 /**
  * 创建Web站点文档
- * @param 参数 
+ * @param 参数
  * {
-    "source_url_list": [
-    "string"
-  ],
-  "selector": "string"
+ "source_url_list": [
+ "string"
+ ],
+ "selector": "string"
+ }
  }
-}
  */
 const postWebDocument: (
   dataset_id: string,
@@ -199,9 +213,9 @@ const postWebDocument: (
 
 /**
  * 导入QA文档
- * @param 参数 
+ * @param 参数
  * file
-}
+ }
  */
 const postQADocument: (
   dataset_id: string,
@@ -323,5 +337,6 @@ export default {
   exportTableTemplate,
   postQADocument,
   postTableDocument,
-  exportDocument
+  exportDocument,
+  batchRefresh
 }
diff --git a/ui/src/views/application/component/AccessSettingDrawer.vue b/ui/src/views/application/component/AccessSettingDrawer.vue
index 8dd7a2e8d..2e9b126fe 100644
--- a/ui/src/views/application/component/AccessSettingDrawer.vue
+++ b/ui/src/views/application/component/AccessSettingDrawer.vue
@@ -28,6 +28,10 @@
           </el-input>
         </el-form-item>
       </template>
+      <div v-if="configType === 'wechat'" class="flex align-center" style="margin-bottom: 8px">
+        <span class="el-form-item__label">是否是订阅号</span>
+        <el-switch v-if="configType === 'wechat'" v-model="form[configType].is_personal" />
+      </div>
 
       <h4 class="title-decoration-1 mb-16">回调地址</h4>
       <el-form-item label="URL" prop="callback_url">
@@ -102,7 +106,14 @@ const {
 } = route as any
 
 const form = reactive<any>({
-  wechat: { app_id: '', app_secret: '', token: '', encoding_aes_key: '', callback_url: '' },
+  wechat: {
+    app_id: '',
+    app_secret: '',
+    token: '',
+    encoding_aes_key: '',
+    is_personal: false,
+    callback_url: ''
+  },
   dingtalk: { client_id: '', client_secret: '', callback_url: '' },
   wecom: {
     app_id: '',
diff --git a/ui/src/views/document/index.vue b/ui/src/views/document/index.vue
index f23c0cc5c..347129096 100644
--- a/ui/src/views/document/index.vue
+++ b/ui/src/views/document/index.vue
@@ -23,6 +23,9 @@
             <el-button @click="openDatasetDialog()" :disabled="multipleSelection.length === 0">
               迁移
             </el-button>
+            <el-button @click="batchRefresh" :disabled="multipleSelection.length === 0">
+              重新向量化
+            </el-button>
             <el-button @click="openBatchEditDocument" :disabled="multipleSelection.length === 0">
               设置
             </el-button>
@@ -538,6 +541,19 @@ function deleteMulDocument() {
   })
 }
 
+function batchRefresh() {
+  const arr: string[] = []
+  multipleSelection.value.map((v) => {
+    if (v) {
+      arr.push(v.id)
+    }
+  })
+  documentApi.batchRefresh(id, arr, loading).then(() => {
+    MsgSuccess('批量重新向量化成功')
+    multipleTableRef.value?.clearSelection()
+  })
+}
+
 function deleteDocument(row: any) {
   MsgConfirm(
     `是否删除文档：${row.name} ?`,