diff --git a/web/i18n/lang/common.en.ts b/web/i18n/lang/common.en.ts
index 1cf6425bb..22eec8d43 100644
--- a/web/i18n/lang/common.en.ts
+++ b/web/i18n/lang/common.en.ts
@@ -50,7 +50,7 @@ const translation = {
         'How much to penalize new tokens based on their existing frequency in the text so far. Decreases the model\'s likelihood to repeat the same line verbatim.',
       maxToken: 'Max token',
       maxTokenTip:
-        'Max tokens generated is 2,048 or 4,000, depending on the model. Prompt and completion share this limit. One token is roughly 1 English character.',
+        'Max tokens depending on the model. Prompt and completion share this limit. One token is roughly 1 English character.',
       setToCurrentModelMaxTokenTip: 'Max token is updated to the maximum token of the current model 4,000.',
     },
     tone: {
diff --git a/web/i18n/lang/common.zh.ts b/web/i18n/lang/common.zh.ts
index 84dc4d804..534abc3e7 100644
--- a/web/i18n/lang/common.zh.ts
+++ b/web/i18n/lang/common.zh.ts
@@ -50,7 +50,7 @@ const translation = {
         'Frequency penalty 是根据重复词在目前文本中的出现频率来对其进行惩罚。正值将不太可能重复常用单词和短语。',
       maxToken: '最大 Token',
       maxTokenTip:
-        '生成的最大令牌数为 2,048 或 4,000，取决于模型。提示和完成共享令牌数限制。一个令牌约等于 1 个英文或 半个中文字符。',
+        '生成的最大令牌数取决于模型。提示和完成共享令牌数限制。一个令牌约等于 1 个英文或 半个中文字符。',
       setToCurrentModelMaxTokenTip: '最大令牌数更新为当前模型最大的令牌数 4,000。',
     },
     tone: {