diff --git a/src/web-ui/src/app/components/NavPanel/components/PersistentFooterActions.tsx b/src/web-ui/src/app/components/NavPanel/components/PersistentFooterActions.tsx
index 7818513e..5589bc1f 100644
--- a/src/web-ui/src/app/components/NavPanel/components/PersistentFooterActions.tsx
+++ b/src/web-ui/src/app/components/NavPanel/components/PersistentFooterActions.tsx
@@ -124,6 +124,7 @@ const PersistentFooterActions: React.FC = () => {
   }, []);
 
   return (
+    <>
     <div className="bitfun-nav-panel__footer">
       <div className="bitfun-nav-panel__footer-left">
         <div className="bitfun-nav-panel__footer-more-wrap">
@@ -244,6 +245,7 @@ const PersistentFooterActions: React.FC = () => {
         onAgree={handleAgreeDisclaimer}
       />
     </Modal>
+    </>
   );
 };
 
diff --git a/src/web-ui/src/infrastructure/config/components/AIModelConfig.scss b/src/web-ui/src/infrastructure/config/components/AIModelConfig.scss
index 4dc97656..e64023c9 100644
--- a/src/web-ui/src/infrastructure/config/components/AIModelConfig.scss
+++ b/src/web-ui/src/infrastructure/config/components/AIModelConfig.scss
@@ -629,16 +629,6 @@
       color: var(--color-accent-400);
     }
 
-    &--image_understanding {
-      background: rgba(168, 85, 247, 0.12);
-      color: rgba(192, 132, 252, 1);
-    }
-
-    &--image_generation {
-      background: rgba(236, 72, 153, 0.12);
-      color: rgba(244, 114, 182, 1);
-    }
-
     &--embedding {
       background: rgba(34, 197, 94, 0.12);
       color: var(--color-success);
@@ -659,11 +649,6 @@
       color: rgba(129, 140, 248, 1);
     }
 
-    &--speech_recognition {
-      background: rgba(168, 85, 247, 0.12);
-      color: rgba(192, 132, 252, 0.8);
-    }
-
     &--more {
       background: var(--element-bg-subtle);
       color: var(--color-text-muted);
diff --git a/src/web-ui/src/infrastructure/config/components/AIModelConfig.tsx b/src/web-ui/src/infrastructure/config/components/AIModelConfig.tsx
index 8d41878d..d69ff438 100644
--- a/src/web-ui/src/infrastructure/config/components/AIModelConfig.tsx
+++ b/src/web-ui/src/infrastructure/config/components/AIModelConfig.tsx
@@ -70,15 +70,8 @@ function uniqModelNames(modelNames: string[]): string[] {
 function getCapabilitiesByCategory(category: ModelCategory): ModelCapability[] {
   switch (category) {
     case 'general_chat':
-      return ['text_chat', 'function_calling'];
-    case 'multimodal':
-      return ['text_chat', 'image_understanding', 'function_calling'];
-    case 'image_generation':
-      return ['image_generation'];
-    case 'speech_recognition':
-      return ['speech_recognition'];
     default:
-      return ['text_chat'];
+      return ['text_chat', 'function_calling'];
   }
 }
 
@@ -200,9 +193,6 @@ const AIModelConfig: React.FC = () => {
   const categoryOptions = useMemo<SelectOption[]>(
     () => [
       { label: t('category.general_chat'), value: 'general_chat' },
-      { label: t('category.multimodal'), value: 'multimodal' },
-      { label: t('category.image_generation'), value: 'image_generation' },
-      { label: t('category.speech_recognition'), value: 'speech_recognition' },
     ],
     [t]
   );
@@ -210,9 +200,6 @@ const AIModelConfig: React.FC = () => {
   const categoryCompactLabels = useMemo<Record<ModelCategory, string>>(
     () => ({
       general_chat: t('categoryIcons.general_chat'),
-      multimodal: t('categoryIcons.multimodal'),
-      image_generation: t('categoryIcons.image_generation'),
-      speech_recognition: t('categoryIcons.speech_recognition'),
     }),
     [t]
   );
@@ -622,11 +609,6 @@ const AIModelConfig: React.FC = () => {
     setIsEditing(true);
   };
 
-  const handleAddModelToExistingProvider = (config: AIModelConfigType) => {
-    handleEditProvider(config);
-  };
-
-  
   const handleEdit = (config: AIModelConfigType) => {
     resetRemoteModelDiscovery();
     setManualModelInput('');
@@ -1391,7 +1373,7 @@ const AIModelConfig: React.FC = () => {
 
             {!isFromTemplate && (
               <>
-                <ConfigPageRow label={`${t('form.modelSelection')} *`} description={editingConfig.category === 'speech_recognition' ? t('form.modelNameHint') : undefined} wide multiline>
+                <ConfigPageRow label={`${t('form.modelSelection')} *`} wide multiline>
                   <div className="bitfun-ai-model-config__control-stack">
                     <div className="bitfun-ai-model-config__model-picker-row">
                       <Select
@@ -1402,7 +1384,7 @@ const AIModelConfig: React.FC = () => {
                             : [String(value)];
                           syncSelectedModelDrafts(nextModelNames, editingConfig, !!editingConfig.id);
                         }}
-                        placeholder={editingConfig.category === 'speech_recognition' ? 'glm-asr' : 'glm-4.7'}
+                        placeholder="glm-4.7"
                         options={availableModelOptions}
                         searchable
                         multiple={!editingConfig.id}
@@ -1732,14 +1714,6 @@ const AIModelConfig: React.FC = () => {
                       >
                         <Edit2 size={14} />
                       </IconButton>
-                      <IconButton
-                        variant="ghost"
-                        size="small"
-                        onClick={() => handleAddModelToExistingProvider(group.models[0])}
-                        tooltip={t('actions.addModel')}
-                      >
-                        <Plus size={14} />
-                      </IconButton>
                     </div>
                   </div>
                   <div className="bitfun-ai-model-config__provider-group-list">
diff --git a/src/web-ui/src/infrastructure/config/components/DefaultModelConfig.tsx b/src/web-ui/src/infrastructure/config/components/DefaultModelConfig.tsx
index a02e38e0..e9c3028a 100644
--- a/src/web-ui/src/infrastructure/config/components/DefaultModelConfig.tsx
+++ b/src/web-ui/src/infrastructure/config/components/DefaultModelConfig.tsx
@@ -13,8 +13,6 @@ import { getProviderDisplayName } from '../services/modelConfigs';
 import type {
   AIModelConfig,
   DefaultModels,
-  OptionalCapabilityModels,
-  OptionalCapabilityType,
 } from '../types';
 import { ConfigPageRow } from './common';
 import { createLogger } from '@/shared/utils/logger';
@@ -22,13 +20,6 @@ import './DefaultModelConfig.scss';
 
 const log = createLogger('DefaultModelConfig');
 
-
-const OPTIONAL_CAPABILITY_TYPES: OptionalCapabilityType[] = [
-  'image_understanding',
-  'image_generation',
-  'speech_recognition'
-];
-
 const normalizeSelectValue = (value: string | number | (string | number)[]): string | number =>
   Array.isArray(value) ? (value[0] ?? '') : value;
 
@@ -50,7 +41,6 @@ export const DefaultModelConfig: React.FC = () => {
   const [loading, setLoading] = useState(true);
   const [models, setModels] = useState<AIModelConfig[]>([]);
   const [defaultModels, setDefaultModels] = useState<DefaultModels>({ primary: null, fast: null });
-  const [optionalCapabilities, setOptionalCapabilities] = useState<OptionalCapabilityModels>({});
 
   const loadData = useCallback(async () => {
     try {
@@ -67,12 +57,6 @@ export const DefaultModelConfig: React.FC = () => {
         primary: defaultModelsConfig?.primary || null,
         fast: defaultModelsConfig?.fast || null,
       });
-
-      setOptionalCapabilities({
-        image_understanding: defaultModelsConfig?.image_understanding,
-        image_generation: defaultModelsConfig?.image_generation,
-        speech_recognition: defaultModelsConfig?.speech_recognition,
-      });
     } catch (error) {
       log.error('Failed to load data', error);
       notificationService.error(t('messages.loadFailed'));
@@ -203,47 +187,7 @@ export const DefaultModelConfig: React.FC = () => {
   };
 
   
-  const handleCapabilityChange = async (capability: OptionalCapabilityType, modelId: string | number) => {
-    const modelIdStr = modelId ? String(modelId) : null;
-    try {
-      const currentConfig = await configManager.getConfig<any>('ai.default_models') || {};
-
-      
-      await configManager.setConfig('ai.default_models', {
-        ...currentConfig,
-        [capability]: modelIdStr || undefined,
-      });
-
-      setOptionalCapabilities(prev => ({
-        ...prev,
-        [capability]: modelIdStr || undefined,
-      }));
-
-      notificationService.success(t('messages.capabilityUpdated'), { duration: 2000 });
-    } catch (error) {
-      log.error('Failed to update capability model', { capability, modelId: modelIdStr, error });
-      notificationService.error(t('messages.updateFailed'));
-    }
-  };
-
-  
   const enabledModels = models.filter(m => m.enabled);
-  
-  
-  const getModelsForCapability = (capability: OptionalCapabilityType): AIModelConfig[] => {
-    return enabledModels.filter(m => {
-      switch (capability) {
-        case 'image_understanding':
-          return m.capabilities?.includes('image_understanding');
-        case 'image_generation':
-          return m.capabilities?.includes('image_generation');
-        case 'speech_recognition':
-          return m.capabilities?.includes('speech_recognition');
-        default:
-          return true;
-      }
-    });
-  };
 
   if (loading) {
     return (
@@ -302,36 +246,6 @@ export const DefaultModelConfig: React.FC = () => {
           size="small"
         />
       </ConfigPageRow>
-
-      {OPTIONAL_CAPABILITY_TYPES.map(capability => {
-        const availableModels = getModelsForCapability(capability);
-        const configuredModelId = optionalCapabilities[capability];
-
-        return (
-          <ConfigPageRow
-            key={capability}
-            label={renderOptionalLabel(t(`optional.capabilities.${capability}.label`))}
-            description={t(`optional.capabilities.${capability}.description`)}
-            align="center"
-          >
-            <Select
-              value={configuredModelId || ''}
-              onChange={(value) => handleCapabilityChange(capability, normalizeSelectValue(value))}
-              placeholder={t('optional.selectModel')}
-              // Allow clearing the selection even when there are no compatible models.
-              disabled={availableModels.length === 0 && !configuredModelId}
-              options={[
-                { label: t('optional.notSet'), value: '' },
-                ...availableModels.map(buildModelOption),
-              ]}
-              renderOption={renderModelOption}
-              renderValue={renderModelValue}
-              className="default-model-config__model-select"
-              size="small"
-            />
-          </ConfigPageRow>
-        );
-      })}
     </div>
   );
 };
diff --git a/src/web-ui/src/infrastructure/config/types/index.ts b/src/web-ui/src/infrastructure/config/types/index.ts
index 658688e7..9eaa67d8 100644
--- a/src/web-ui/src/infrastructure/config/types/index.ts
+++ b/src/web-ui/src/infrastructure/config/types/index.ts
@@ -68,16 +68,10 @@ export interface AIExperienceConfig {
 
 export type ModelCapability =
   | 'text_chat'
-  | 'image_understanding'
-  | 'image_generation'
-  | 'function_calling'
-  | 'speech_recognition';
+  | 'function_calling';
 
 export type ModelCategory = 
-  | 'general_chat'
-  | 'multimodal'
-  | 'image_generation'
-  | 'speech_recognition';
+  | 'general_chat';
 
 export interface ModelMetadata {
   category: ModelCategory;
@@ -87,28 +81,13 @@ export interface ModelMetadata {
 }
 
 
-export const CAPABILITY_LABELS: Record<ModelCapability, string> = {
-  text_chat: t('settings/ai-model:capabilities.text_chat'),
-  image_understanding: t('settings/ai-model:capabilities.image_understanding'),
-  image_generation: t('settings/ai-model:capabilities.image_generation'),
-  function_calling: t('settings/ai-model:capabilities.function_calling'),
-  speech_recognition: t('settings/ai-model:capabilities.speech_recognition')
-};
-
-
 export const CATEGORY_LABELS: Record<ModelCategory, string> = {
-  general_chat: t('settings/ai-model:category.general_chat'),
-  multimodal: t('settings/ai-model:category.multimodal'),
-  image_generation: t('settings/ai-model:category.image_generation'),
-  speech_recognition: t('settings/ai-model:category.speech_recognition')
+  general_chat: t('settings/ai-model:category.general_chat')
 };
 
 
 export const CATEGORY_ICONS: Record<ModelCategory, string> = {
-  general_chat: t('settings/ai-model:categoryIcons.general_chat'),
-  multimodal: t('settings/ai-model:categoryIcons.multimodal'),
-  image_generation: t('settings/ai-model:categoryIcons.image_generation'),
-  speech_recognition: t('settings/ai-model:categoryIcons.speech_recognition')
+  general_chat: t('settings/ai-model:categoryIcons.general_chat')
 };
 
 
@@ -168,10 +147,6 @@ export interface DefaultModelsConfig {
   primary?: string | null;
    
   fast?: string | null;
-   
-  image_understanding?: string | null;
-   
-  speech_recognition?: string | null;
 }
 
 export interface AIConfig {
@@ -582,16 +557,4 @@ export interface DefaultModels {
 
  
 export interface OptionalCapabilityModels {
-   
-  image_understanding?: string;
-   
-  image_generation?: string;
-   
-  speech_recognition?: string;
 }
-
- 
-export type OptionalCapabilityType =
-  | 'image_understanding'
-  | 'image_generation'
-  | 'speech_recognition';