diff --git a/package.json b/package.json
index 46d27ca7..21c16066 100644
--- a/package.json
+++ b/package.json
@@ -1,6 +1,6 @@
 {
   "name": "cherry-studio",
-  "version": "0.4.9",
+  "version": "0.5.0",
   "description": "A powerful AI assistant for producer.",
   "main": "./out/main/index.js",
   "author": "kangfenmao@qq.com",
diff --git a/src/main/index.ts b/src/main/index.ts
index 3f108c6e..885d74e7 100644
--- a/src/main/index.ts
+++ b/src/main/index.ts
@@ -26,7 +26,7 @@ function createWindow() {
     width: mainWindowState.width,
     height: mainWindowState.height,
     minWidth: 1080,
-    minHeight: 500,
+    minHeight: 600,
     show: true,
     autoHideMenuBar: true,
     transparent: process.platform === 'darwin',
diff --git a/src/renderer/src/pages/home/components/sidebar/SettingsTab.tsx b/src/renderer/src/pages/home/components/sidebar/SettingsTab.tsx
index 9172878e..58554b45 100644
--- a/src/renderer/src/pages/home/components/sidebar/SettingsTab.tsx
+++ b/src/renderer/src/pages/home/components/sidebar/SettingsTab.tsx
@@ -33,7 +33,6 @@ const SettingsTab: FC<Props> = (props) => {
     debounce(
       (settings: Partial<AssistantSettings>) => {
         updateAssistantSettings({
-          ...assistant.settings,
           temperature: settings.temperature ?? temperature,
           contextCount: settings.contextCount ?? contextCount,
           enableMaxTokens: settings.enableMaxTokens ?? enableMaxTokens,
@@ -41,12 +40,9 @@ const SettingsTab: FC<Props> = (props) => {
         })
       },
       1000,
-      {
-        leading: false,
-        trailing: true
-      }
+      { leading: true, trailing: false }
     ),
-    []
+    [temperature, contextCount, enableMaxTokens, maxTokens]
   )
 
   const onTemperatureChange = (value) => {
@@ -255,7 +251,7 @@ const InputNumberic = styled(InputNumber)`
 const Label = styled.p`
   margin: 0;
   font-size: 12px;
-  font-weight: bold;
+  font-weight: 600;
   margin-right: 8px;
 `
 
diff --git a/src/renderer/src/pages/settings/AssistantSettings.tsx b/src/renderer/src/pages/settings/AssistantSettings.tsx
index 00660436..854881bb 100644
--- a/src/renderer/src/pages/settings/AssistantSettings.tsx
+++ b/src/renderer/src/pages/settings/AssistantSettings.tsx
@@ -38,7 +38,7 @@ const AssistantSettings: FC = () => {
       1000,
       { leading: false, trailing: true }
     ),
-    []
+    [temperature, contextCount, enableMaxTokens, maxTokens]
   )
 
   const onTemperatureChange = (value) => {
diff --git a/src/renderer/src/services/ProviderSDK.ts b/src/renderer/src/services/ProviderSDK.ts
index 22a0aaf2..fe17ae41 100644
--- a/src/renderer/src/services/ProviderSDK.ts
+++ b/src/renderer/src/services/ProviderSDK.ts
@@ -3,12 +3,12 @@ import { MessageCreateParamsNonStreaming, MessageParam } from '@anthropic-ai/sdk
 import { DEFAULT_MAX_TOKENS } from '@renderer/config/constant'
 import { getOllamaKeepAliveTime } from '@renderer/hooks/useOllama'
 import { Assistant, Message, Provider, Suggestion } from '@renderer/types'
-import { getAssistantSettings, removeQuotes } from '@renderer/utils'
+import { removeQuotes } from '@renderer/utils'
 import { sum, takeRight } from 'lodash'
 import OpenAI from 'openai'
 import { ChatCompletionCreateParamsNonStreaming, ChatCompletionMessageParam } from 'openai/resources'
 
-import { getAssistantMaxTokens, getDefaultModel, getTopNamingModel } from './assistant'
+import { getAssistantSettings, getDefaultModel, getTopNamingModel } from './assistant'
 import { EVENT_NAMES } from './event'
 
 export default class ProviderSDK {
@@ -39,7 +39,7 @@ export default class ProviderSDK {
   ) {
     const defaultModel = getDefaultModel()
     const model = assistant.model || defaultModel
-    const { contextCount } = getAssistantSettings(assistant)
+    const { contextCount, maxTokens } = getAssistantSettings(assistant)
 
     const systemMessage = assistant.prompt ? { role: 'system', content: assistant.prompt } : undefined
 
@@ -53,7 +53,7 @@ export default class ProviderSDK {
         .stream({
           model: model.id,
           messages: [systemMessage, ...userMessages].filter(Boolean) as MessageParam[],
-          max_tokens: getAssistantMaxTokens(assistant) || DEFAULT_MAX_TOKENS,
+          max_tokens: maxTokens || DEFAULT_MAX_TOKENS,
           temperature: assistant?.settings?.temperature
         })
         .on('text', (text) => onChunk({ text: text || '' }))
@@ -73,7 +73,7 @@ export default class ProviderSDK {
         messages: [systemMessage, ...userMessages].filter(Boolean) as ChatCompletionMessageParam[],
         stream: true,
         temperature: assistant?.settings?.temperature,
-        max_tokens: getAssistantMaxTokens(assistant),
+        max_tokens: maxTokens,
         keep_alive: this.keepAliveTime
       })
       for await (const chunk of stream) {
diff --git a/src/renderer/src/services/assistant.ts b/src/renderer/src/services/assistant.ts
index 2b8614fd..b85ca8ca 100644
--- a/src/renderer/src/services/assistant.ts
+++ b/src/renderer/src/services/assistant.ts
@@ -1,9 +1,9 @@
-import { DEFAULT_MAX_TOKENS } from '@renderer/config/constant'
+import { DEFAULT_CONEXTCOUNT, DEFAULT_MAX_TOKENS, DEFAULT_TEMPERATURE } from '@renderer/config/constant'
 import i18n from '@renderer/i18n'
 import store from '@renderer/store'
 import { updateAgent } from '@renderer/store/agents'
 import { updateAssistant } from '@renderer/store/assistants'
-import { Agent, Assistant, Model, Provider, Topic } from '@renderer/types'
+import { Agent, Assistant, AssistantSettings, Model, Provider, Topic } from '@renderer/types'
 import { getLeadingEmoji, removeLeadingEmoji, uuid } from '@renderer/utils'
 
 export function getDefaultAssistant(): Assistant {
@@ -57,16 +57,25 @@ export function getProviderByModelId(modelId?: string) {
   return providers.find((p) => p.models.find((m) => m.id === _modelId)) as Provider
 }
 
-export function getAssistantMaxTokens(assistant: Assistant) {
-  if (assistant.settings?.enableMaxTokens) {
-    const maxTokens = assistant.settings.maxTokens
-    if (typeof maxTokens === 'number') {
-      return maxTokens > 100 ? maxTokens : DEFAULT_MAX_TOKENS
+export const getAssistantSettings = (assistant: Assistant): AssistantSettings => {
+  const contextCount = assistant?.settings?.contextCount ?? DEFAULT_CONEXTCOUNT
+  const getAssistantMaxTokens = () => {
+    if (assistant.settings?.enableMaxTokens) {
+      const maxTokens = assistant.settings.maxTokens
+      if (typeof maxTokens === 'number') {
+        return maxTokens > 100 ? maxTokens : DEFAULT_MAX_TOKENS
+      }
+      return DEFAULT_MAX_TOKENS
     }
-    return DEFAULT_MAX_TOKENS
+    return undefined
   }
 
-  return undefined
+  return {
+    contextCount: contextCount === 20 ? 100000 : contextCount,
+    temperature: assistant?.settings?.temperature ?? DEFAULT_TEMPERATURE,
+    enableMaxTokens: assistant?.settings?.enableMaxTokens ?? false,
+    maxTokens: getAssistantMaxTokens()
+  }
 }
 
 export function covertAgentToAssistant(agent: Agent): Assistant {
diff --git a/src/renderer/src/services/messages.ts b/src/renderer/src/services/messages.ts
index 84166593..50497872 100644
--- a/src/renderer/src/services/messages.ts
+++ b/src/renderer/src/services/messages.ts
@@ -1,8 +1,9 @@
 import { Assistant, Message } from '@renderer/types'
-import { getAssistantSettings } from '@renderer/utils'
 import { GPTTokens } from 'gpt-tokens'
 import { takeRight } from 'lodash'
 
+import { getAssistantSettings } from './assistant'
+
 export const filterAtMessages = (messages: Message[]) => {
   return messages.filter((message) => message.type !== '@')
 }
diff --git a/src/renderer/src/types/index.ts b/src/renderer/src/types/index.ts
index b2468a49..122617d9 100644
--- a/src/renderer/src/types/index.ts
+++ b/src/renderer/src/types/index.ts
@@ -14,7 +14,7 @@ export type Assistant = {
 export type AssistantSettings = {
   contextCount: number
   temperature: number
-  maxTokens: number
+  maxTokens: number | undefined
   enableMaxTokens: boolean
 }
 
diff --git a/src/renderer/src/utils/index.ts b/src/renderer/src/utils/index.ts
index a9d0712d..1c0ca018 100644
--- a/src/renderer/src/utils/index.ts
+++ b/src/renderer/src/utils/index.ts
@@ -1,5 +1,4 @@
-import { DEFAULT_CONEXTCOUNT, DEFAULT_TEMPERATURE } from '@renderer/config/constant'
-import { Assistant, AssistantSettings, Model } from '@renderer/types'
+import { Model } from '@renderer/types'
 import imageCompression from 'browser-image-compression'
 import { v4 as uuidv4 } from 'uuid'
 
@@ -177,14 +176,6 @@ export function getFirstCharacter(str) {
   }
 }
 
-export const getAssistantSettings = (assistant: Assistant): AssistantSettings => {
-  const contextCount = assistant?.settings?.contextCount ?? DEFAULT_CONEXTCOUNT
-  return {
-    contextCount: contextCount === 20 ? 100000 : contextCount,
-    temperature: assistant?.settings?.temperature ?? DEFAULT_TEMPERATURE
-  }
-}
-
 /**
  * is valid proxy url
  * @param url proxy url