chore: bump version to 1.7.0-sora.2

- Add video download logic with progress tracking in VideoPanel
- Reset load state when video changes in VideoViewer
- Improve video player styling and loading state handling
- Add file upload and metadata handling for downloaded videos

package.json

@@ -1,6 +1,6 @@
 {
   "name": "CherryStudio",
-  "version": "1.7.0-alpha.5",
+  "version": "1.7.0-sora.2",
   "private": true,
   "description": "A powerful AI assistant for producer.",
   "main": "./out/main/index.js",
@@ -124,6 +124,7 @@
     "@cherrystudio/embedjs-ollama": "^0.1.31",
     "@cherrystudio/embedjs-openai": "^0.1.31",
     "@cherrystudio/extension-table-plus": "workspace:^",
+    "@cherrystudio/openai": "6.3.0-fork.1",
     "@dnd-kit/core": "^6.3.1",
     "@dnd-kit/modifiers": "^9.0.0",
     "@dnd-kit/sortable": "^10.0.0",
@@ -295,7 +296,6 @@
     "motion": "^12.10.5",
     "notion-helper": "^1.3.22",
     "npx-scope-finder": "^1.2.0",
-    "openai": "patch:openai@npm%3A5.12.2#~/.yarn/patches/openai-npm-5.12.2-30b075401c.patch",
     "oxlint": "^1.22.0",
     "oxlint-tsgolint": "^0.2.0",
     "p-queue": "^8.1.0",
@@ -375,8 +375,8 @@
     "file-stream-rotator@npm:^0.6.1": "patch:file-stream-rotator@npm%3A0.6.1#~/.yarn/patches/file-stream-rotator-npm-0.6.1-eab45fb13d.patch",
     "libsql@npm:^0.4.4": "patch:libsql@npm%3A0.4.7#~/.yarn/patches/libsql-npm-0.4.7-444e260fb1.patch",
     "node-abi": "4.12.0",
-    "openai@npm:^4.77.0": "patch:openai@npm%3A5.12.2#~/.yarn/patches/openai-npm-5.12.2-30b075401c.patch",
+    "openai@npm:^4.77.0": "npm:@cherrystudio/openai@6.3.0-fork.1",
-    "openai@npm:^4.87.3": "patch:openai@npm%3A5.12.2#~/.yarn/patches/openai-npm-5.12.2-30b075401c.patch",
+    "openai@npm:^4.87.3": "npm:@cherrystudio/openai@6.3.0-fork.1",
     "pdf-parse@npm:1.1.1": "patch:pdf-parse@npm%3A1.1.1#~/.yarn/patches/pdf-parse-npm-1.1.1-04a6109b2a.patch",
     "pkce-challenge@npm:^4.1.0": "patch:pkce-challenge@npm%3A4.1.0#~/.yarn/patches/pkce-challenge-npm-4.1.0-fbc51695a3.patch",
     "undici": "6.21.2",

scripts/auto-translate-i18n.ts

@@ -2,9 +2,9 @@
  * 该脚本用于少量自动翻译所有baseLocale以外的文本。待翻译文案必须以[to be translated]开头
  *
  */
+import OpenAI from '@cherrystudio/openai'
 import cliProgress from 'cli-progress'
 import * as fs from 'fs'
-import OpenAI from 'openai'
 import * as path from 'path'
 
 const localesDir = path.join(__dirname, '../src/renderer/src/i18n/locales')

scripts/update-i18n.ts

@@ -4,9 +4,9 @@
  * API_KEY=sk-xxxx BASE_URL=xxxx MODEL=xxxx ts-node scripts/update-i18n.ts
  */
 
+import OpenAI from '@cherrystudio/openai'
 import cliProgress from 'cli-progress'
 import fs from 'fs'
-import OpenAI from 'openai'
 
 type I18NValue = string | { [key: string]: I18NValue }
 type I18N = { [key: string]: I18NValue }

src/main/apiServer/routes/chat.ts

@@ -1,5 +1,5 @@
+import { ChatCompletionCreateParams } from '@cherrystudio/openai/resources'
 import express, { Request, Response } from 'express'
-import { ChatCompletionCreateParams } from 'openai/resources'
 
 import { loggerService } from '../../services/LoggerService'
 import {

src/main/apiServer/services/chat-completion.ts

@@ -1,6 +1,6 @@
+import OpenAI from '@cherrystudio/openai'
+import { ChatCompletionCreateParams, ChatCompletionCreateParamsStreaming } from '@cherrystudio/openai/resources'
 import { Provider } from '@types'
-import OpenAI from 'openai'
-import { ChatCompletionCreateParams, ChatCompletionCreateParamsStreaming } from 'openai/resources'
 
 import { loggerService } from '../../services/LoggerService'
 import { ModelValidationError, validateModelId } from '../utils'

src/main/services/remotefile/OpenAIService.ts

@@ -1,8 +1,8 @@
+import OpenAI from '@cherrystudio/openai'
 import { loggerService } from '@logger'
 import { fileStorage } from '@main/services/FileStorage'
 import { FileListResponse, FileMetadata, FileUploadResponse, Provider } from '@types'
 import * as fs from 'fs'
-import OpenAI from 'openai'
 
 import { CacheService } from '../CacheService'
 import { BaseFileService } from './BaseFileService'

src/renderer/src/Router.tsx

@@ -20,6 +20,7 @@ import PaintingsRoutePage from './pages/paintings/PaintingsRoutePage'
 import SettingsPage from './pages/settings/SettingsPage'
 import AssistantPresetsPage from './pages/store/assistants/presets/AssistantPresetsPage'
 import TranslatePage from './pages/translate/TranslatePage'
+import { VideoPage } from './pages/video/VideoPage'
 
 const Router: FC = () => {
   const { navbarPosition } = useNavbarPosition()
@@ -40,6 +41,7 @@ const Router: FC = () => {
           <Route path="/code" element={<CodeToolsPage />} />
           <Route path="/settings/*" element={<SettingsPage />} />
           <Route path="/launchpad" element={<LaunchpadPage />} />
+          <Route path="/video" element={<VideoPage />} />
         </Routes>
       </ErrorBoundary>
     )

src/renderer/src/aiCore/index_new.ts

@@ -12,8 +12,15 @@ import { loggerService } from '@logger'
 import { getEnableDeveloperMode } from '@renderer/hooks/useSettings'
 import { addSpan, endSpan } from '@renderer/services/SpanManagerService'
 import { StartSpanParams } from '@renderer/trace/types/ModelSpanEntity'
-import type { Assistant, GenerateImageParams, Model, Provider } from '@renderer/types'
+import type { Assistant, GenerateImageParams, Model, Provider, RetrieveVideoContentParams } from '@renderer/types'
 import type { AiSdkModel, StreamTextParams } from '@renderer/types/aiCoreTypes'
+import {
+  CreateVideoParams,
+  CreateVideoResult,
+  RetrieveVideoContentResult,
+  RetrieveVideoParams,
+  RetrieveVideoResult
+} from '@renderer/types/video'
 import { buildClaudeCodeSystemModelMessage } from '@shared/anthropic'
 import { type ImageModel, type LanguageModel, type Provider as AiSdkProvider, wrapLanguageModel } from 'ai'
 
@@ -500,6 +507,27 @@ export default class ModernAiProvider {
     return images
   }
 
+  /**
+   * We manually implement this method before aisdk supports it well
+   */
+  public async createVideo(params: CreateVideoParams): Promise<CreateVideoResult> {
+    return this.legacyProvider.createVideo(params)
+  }
+
+  /**
+   * We manually implement this method before aisdk supports it well
+   */
+  public async retrieveVideo(params: RetrieveVideoParams): Promise<RetrieveVideoResult> {
+    return this.legacyProvider.retrieveVideo(params)
+  }
+
+  /**
+   * We manually implement this method before aisdk supports it well
+   */
+  public async retrieveVideoContent(params: RetrieveVideoContentParams): Promise<RetrieveVideoContentResult> {
+    return this.legacyProvider.retrieveVideoContent(params)
+  }
+
   public getBaseURL(): string {
     return this.legacyProvider.getBaseURL()
   }

src/renderer/src/aiCore/legacy/clients/cherryai/CherryAiAPIClient.ts

@@ -1,6 +1,6 @@
+import OpenAI from '@cherrystudio/openai'
 import { Provider } from '@renderer/types'
 import { OpenAISdkParams, OpenAISdkRawOutput } from '@renderer/types/sdk'
-import OpenAI from 'openai'
 
 import { OpenAIAPIClient } from '../openai/OpenAIApiClient'
 

src/renderer/src/aiCore/legacy/clients/openai/OpenAIApiClient.ts

@@ -1,3 +1,9 @@
+import OpenAI, { AzureOpenAI } from '@cherrystudio/openai'
+import {
+  ChatCompletionContentPart,
+  ChatCompletionContentPartRefusal,
+  ChatCompletionTool
+} from '@cherrystudio/openai/resources'
 import { loggerService } from '@logger'
 import { DEFAULT_MAX_TOKENS } from '@renderer/config/constant'
 import {
@@ -78,8 +84,6 @@ import {
 } from '@renderer/utils/mcp-tools'
 import { findFileBlocks, findImageBlocks } from '@renderer/utils/messageUtils/find'
 import { t } from 'i18next'
-import OpenAI, { AzureOpenAI } from 'openai'
-import { ChatCompletionContentPart, ChatCompletionContentPartRefusal, ChatCompletionTool } from 'openai/resources'
 
 import { GenericChunk } from '../../middleware/schemas'
 import { RequestTransformer, ResponseChunkTransformer, ResponseChunkTransformerContext } from '../types'

src/renderer/src/aiCore/legacy/clients/openai/OpenAIBaseClient.ts

@@ -1,3 +1,4 @@
+import OpenAI, { AzureOpenAI } from '@cherrystudio/openai'
 import { loggerService } from '@logger'
 import {
   isClaudeReasoningModel,
@@ -24,7 +25,6 @@ import {
   ReasoningEffortOptionalParams
 } from '@renderer/types/sdk'
 import { formatApiHost } from '@renderer/utils/api'
-import OpenAI, { AzureOpenAI } from 'openai'
 
 import { BaseApiClient } from '../BaseApiClient'
 

src/renderer/src/aiCore/legacy/clients/openai/OpenAIResponseAPIClient.ts

@@ -1,3 +1,5 @@
+import OpenAI, { AzureOpenAI } from '@cherrystudio/openai'
+import { ResponseInput } from '@cherrystudio/openai/resources/responses/responses'
 import { loggerService } from '@logger'
 import { GenericChunk } from '@renderer/aiCore/legacy/middleware/schemas'
 import { CompletionsContext } from '@renderer/aiCore/legacy/middleware/types'
@@ -34,6 +36,7 @@ import {
   OpenAIResponseSdkTool,
   OpenAIResponseSdkToolCall
 } from '@renderer/types/sdk'
+import { CreateVideoParams, RetrieveVideoContentParams, RetrieveVideoParams } from '@renderer/types/video'
 import { addImageFileToContents } from '@renderer/utils/formats'
 import {
   isSupportedToolUse,
@@ -45,8 +48,6 @@ import { findFileBlocks, findImageBlocks } from '@renderer/utils/messageUtils/fi
 import { MB } from '@shared/config/constant'
 import { t } from 'i18next'
 import { isEmpty } from 'lodash'
-import OpenAI, { AzureOpenAI } from 'openai'
-import { ResponseInput } from 'openai/resources/responses/responses'
 
 import { RequestTransformer, ResponseChunkTransformer } from '../types'
 import { OpenAIAPIClient } from './OpenAIApiClient'
@@ -152,6 +153,21 @@ export class OpenAIResponseAPIClient extends OpenAIBaseClient<
     return await sdk.responses.create(payload, options)
   }
 
+  public async createVideo(params: CreateVideoParams): Promise<OpenAI.Videos.Video> {
+    const sdk = await this.getSdkInstance()
+    return sdk.videos.create(params.params, params.options)
+  }
+
+  public async retrieveVideo(params: RetrieveVideoParams): Promise<OpenAI.Videos.Video> {
+    const sdk = await this.getSdkInstance()
+    return sdk.videos.retrieve(params.videoId, params.options)
+  }
+
+  public async retrieveVideoContent(params: RetrieveVideoContentParams): Promise<Response> {
+    const sdk = await this.getSdkInstance()
+    return sdk.videos.downloadContent(params.videoId, params.query, params.options)
+  }
+
   private async handlePdfFile(file: FileMetadata): Promise<OpenAI.Responses.ResponseInputFile | undefined> {
     if (file.size > 32 * MB) return undefined
     try {
@@ -343,7 +359,14 @@ export class OpenAIResponseAPIClient extends OpenAIBaseClient<
     }
     switch (message.type) {
       case 'function_call_output':
-        sum += estimateTextTokens(message.output)
+        if (typeof message.output === 'string') {
+          sum += estimateTextTokens(message.output)
+        } else {
+          sum += message.output
+            .filter((item) => item.type === 'input_text')
+            .map((item) => estimateTextTokens(item.text))
+            .reduce((prev, cur) => prev + cur, 0)
+        }
         break
       case 'function_call':
         sum += estimateTextTokens(message.arguments)

src/renderer/src/aiCore/legacy/clients/ovms/OVMSClient.ts

@@ -1,7 +1,7 @@
+import OpenAI from '@cherrystudio/openai'
 import { loggerService } from '@logger'
 import { isSupportedModel } from '@renderer/config/models'
 import { objectKeys, Provider } from '@renderer/types'
-import OpenAI from 'openai'
 
 import { OpenAIAPIClient } from '../openai/OpenAIApiClient'
 

src/renderer/src/aiCore/legacy/clients/ppio/PPIOAPIClient.ts

@@ -1,7 +1,7 @@
+import OpenAI from '@cherrystudio/openai'
 import { loggerService } from '@logger'
 import { isSupportedModel } from '@renderer/config/models'
 import { Model, Provider } from '@renderer/types'
-import OpenAI from 'openai'
 
 import { OpenAIAPIClient } from '../openai/OpenAIApiClient'
 

src/renderer/src/aiCore/legacy/clients/types.ts

@@ -1,4 +1,5 @@
 import Anthropic from '@anthropic-ai/sdk'
+import OpenAI from '@cherrystudio/openai'
 import { Assistant, MCPTool, MCPToolResponse, Model, ToolCallResponse } from '@renderer/types'
 import { Provider } from '@renderer/types'
 import {
@@ -13,7 +14,6 @@ import {
   SdkTool,
   SdkToolCall
 } from '@renderer/types/sdk'
-import OpenAI from 'openai'
 
 import { CompletionsParams, GenericChunk } from '../middleware/schemas'
 import { CompletionsContext } from '../middleware/types'

src/renderer/src/aiCore/legacy/clients/zhipu/ZhipuAPIClient.ts

@@ -1,7 +1,7 @@
+import OpenAI from '@cherrystudio/openai'
 import { loggerService } from '@logger'
 import { Provider } from '@renderer/types'
 import { GenerateImageParams } from '@renderer/types'
-import OpenAI from 'openai'
 
 import { OpenAIAPIClient } from '../openai/OpenAIApiClient'
 

src/renderer/src/aiCore/legacy/index.ts

@@ -5,8 +5,15 @@ import { isDedicatedImageGenerationModel, isFunctionCallingModel } from '@render
 import { getProviderByModel } from '@renderer/services/AssistantService'
 import { withSpanResult } from '@renderer/services/SpanManagerService'
 import { StartSpanParams } from '@renderer/trace/types/ModelSpanEntity'
-import type { GenerateImageParams, Model, Provider } from '@renderer/types'
+import type { GenerateImageParams, Model, Provider, RetrieveVideoContentParams } from '@renderer/types'
 import type { RequestOptions, SdkModel } from '@renderer/types/sdk'
+import {
+  CreateVideoParams,
+  CreateVideoResult,
+  RetrieveVideoContentResult,
+  RetrieveVideoParams,
+  RetrieveVideoResult
+} from '@renderer/types/video'
 import { isSupportedToolUse } from '@renderer/utils/mcp-tools'
 
 import { AihubmixAPIClient } from './clients/aihubmix/AihubmixAPIClient'
@@ -179,6 +186,42 @@ export default class AiProvider {
     return this.apiClient.generateImage(params)
   }
 
+  public async createVideo(params: CreateVideoParams): Promise<CreateVideoResult> {
+    if (this.apiClient instanceof OpenAIResponseAPIClient && params.type === 'openai') {
+      const video = await this.apiClient.createVideo(params)
+      return {
+        type: 'openai',
+        video
+      }
+    } else {
+      throw new Error('Video generation is not supported by this provider')
+    }
+  }
+
+  public async retrieveVideo(params: RetrieveVideoParams): Promise<RetrieveVideoResult> {
+    if (this.apiClient instanceof OpenAIResponseAPIClient && params.type === 'openai') {
+      const video = await this.apiClient.retrieveVideo(params)
+      return {
+        type: 'openai',
+        video
+      }
+    } else {
+      throw new Error('Video generation is not supported by this provider')
+    }
+  }
+
+  public async retrieveVideoContent(params: RetrieveVideoContentParams): Promise<RetrieveVideoContentResult> {
+    if (this.apiClient instanceof OpenAIResponseAPIClient && params.type === 'openai') {
+      const response = await this.apiClient.retrieveVideoContent(params)
+      return {
+        type: 'openai',
+        response
+      }
+    } else {
+      throw new Error('Video generation is not supported by this provider')
+    }
+  }
+
   public getBaseURL(): string {
     return this.apiClient.getBaseURL()
   }

src/renderer/src/aiCore/legacy/middleware/feat/ImageGenerationMiddleware.ts

@@ -1,10 +1,10 @@
+import OpenAI from '@cherrystudio/openai'
+import { toFile } from '@cherrystudio/openai/uploads'
 import { isDedicatedImageGenerationModel } from '@renderer/config/models'
 import FileManager from '@renderer/services/FileManager'
 import { ChunkType } from '@renderer/types/chunk'
 import { findImageBlocks, getMainTextContent } from '@renderer/utils/messageUtils/find'
 import { defaultTimeout } from '@shared/config/constant'
-import OpenAI from 'openai'
-import { toFile } from 'openai/uploads'
 
 import { BaseApiClient } from '../../clients/BaseApiClient'
 import { CompletionsParams, CompletionsResult, GenericChunk } from '../schemas'

src/renderer/src/aiCore/prepareParams/fileProcessor.ts

@@ -3,6 +3,7 @@
  * 处理文件内容提取、文件格式转换、文件上传等逻辑
  */
 
+import type OpenAI from '@cherrystudio/openai'
 import { loggerService } from '@logger'
 import { getProviderByModel } from '@renderer/services/AssistantService'
 import type { FileMetadata, Message, Model } from '@renderer/types'
@@ -10,7 +11,6 @@ import { FileTypes } from '@renderer/types'
 import { FileMessageBlock } from '@renderer/types/newMessage'
 import { findFileBlocks } from '@renderer/utils/messageUtils/find'
 import type { FilePart, TextPart } from 'ai'
-import type OpenAI from 'openai'
 
 import { getAiSdkProviderId } from '../provider/factory'
 import { getFileSizeLimit, supportsImageInput, supportsLargeFileUpload, supportsPdfInput } from './modelCapabilities'

src/renderer/src/components/Tab/TabContainer.tsx

@@ -32,6 +32,7 @@ import {
   Sparkle,
   Sun,
   Terminal,
+  Video,
   X
 } from 'lucide-react'
 import { useCallback, useEffect, useMemo } from 'react'
@@ -106,6 +107,8 @@ const getTabIcon = (
       return <Settings size={14} />
     case 'code':
       return <Terminal size={14} />
+    case 'video':
+      return <Video size={14} />
     default:
       return null
   }

src/renderer/src/components/ToastPortal.tsx

@@ -23,7 +23,8 @@ export const ToastPortal = () => {
         timeout: 3000,
         classNames: {
           // This setting causes the 'hero-toast' class to be applied twice to the toast element. This is weird and I don't know why, but it works.
-          base: 'hero-toast'
+          // `w-auto` would not overwrite default style, which set the width to a fixed value and causes text overflow.
+          base: 'hero-toast w-auto! max-w-[50vw]'
         }
       }}
     />,

src/renderer/src/components/app/Sidebar.tsx

@@ -26,7 +26,8 @@ import {
   Palette,
   Settings,
   Sparkle,
-  Sun
+  Sun,
+  Video
 } from 'lucide-react'
 import { FC } from 'react'
 import { useTranslation } from 'react-i18next'
@@ -139,7 +140,8 @@ const MainMenus: FC = () => {
     knowledge: <FileSearch size={18} className="icon" />,
     files: <Folder size={18} className="icon" />,
     notes: <NotepadText size={18} className="icon" />,
-    code_tools: <Code size={18} className="icon" />
+    code_tools: <Code size={18} className="icon" />,
+    video: <Video size={18} className="icon" />
   }
 
   const pathMap = {
@@ -151,7 +153,8 @@ const MainMenus: FC = () => {
     knowledge: '/knowledge',
     files: '/files',
     code_tools: '/code',
-    notes: '/notes'
+    notes: '/notes',
+    video: '/video'
   }
 
   return sidebarIcons.visible.map((icon) => {

src/renderer/src/config/models/utils.ts

@@ -1,6 +1,6 @@
+import OpenAI from '@cherrystudio/openai'
 import { Model } from '@renderer/types'
 import { getLowerBaseModelName } from '@renderer/utils'
-import OpenAI from 'openai'
 
 import { WEB_SEARCH_PROMPT_FOR_OPENROUTER } from '../prompts'
 import { getWebSearchTools } from '../tools'

src/renderer/src/config/models/video.ts

@@ -0,0 +1,149 @@
+import { SystemProviderId, Video } from '@renderer/types'
+
+// Hard-encoded for now. We may implement a function to filter video generation model from provider.models.
+export const videoModelsMap = {
+  openai: ['sora-2', 'sora-2-pro'] as const
+} as const satisfies Partial<Record<SystemProviderId, string[]>>
+
+// Mock data for testing
+export const mockVideos: Video[] = [
+  {
+    id: '1',
+    type: 'openai',
+    status: 'downloaded',
+    prompt: 'A beautiful sunset over the ocean with waves crashing',
+    thumbnail: 'https://picsum.photos/200/200?random=1',
+    fileId: 'file-001',
+    providerId: 'openai',
+    name: 'video-001',
+    metadata: {
+      id: 'video-001',
+      object: 'video',
+      created_at: Math.floor(Date.now() / 1000),
+      completed_at: Math.floor(Date.now() / 1000),
+      expires_at: null,
+      error: null,
+      model: 'sora-2',
+      progress: 100,
+      remixed_from_video_id: null,
+      seconds: '4',
+      size: '1280x720',
+      status: 'completed'
+    }
+  },
+  {
+    id: '2',
+    type: 'openai',
+    status: 'in_progress',
+    prompt: 'A cat playing with a ball of yarn in slow motion',
+    progress: 65,
+    providerId: 'openai',
+    name: 'video-002',
+    metadata: {
+      id: 'video-002',
+      object: 'video',
+      created_at: Math.floor(Date.now() / 1000),
+      completed_at: null,
+      expires_at: null,
+      error: null,
+      model: 'sora-2-pro',
+      progress: 65,
+      remixed_from_video_id: null,
+      seconds: '8',
+      size: '1792x1024',
+      status: 'in_progress'
+    }
+  },
+  {
+    id: '3',
+    type: 'openai',
+    status: 'queued',
+    prompt: 'Time-lapse of flowers blooming in a garden',
+    providerId: 'openai',
+    name: 'video-003',
+    metadata: {
+      id: 'video-003',
+      object: 'video',
+      created_at: Math.floor(Date.now() / 1000),
+      completed_at: null,
+      expires_at: null,
+      error: null,
+      model: 'sora-2',
+      progress: 0,
+      remixed_from_video_id: null,
+      seconds: '12',
+      size: '1280x720',
+      status: 'queued'
+    }
+  },
+  {
+    id: '4',
+    type: 'openai',
+    prompt: 'Birds flying in formation against blue sky',
+    status: 'downloading',
+    progress: 80,
+    thumbnail: 'https://picsum.photos/200/200?random=4',
+    providerId: 'openai',
+    name: 'video-004',
+    metadata: {
+      id: 'video-004',
+      object: 'video',
+      created_at: Math.floor(Date.now() / 1000),
+      completed_at: Math.floor(Date.now() / 1000),
+      expires_at: null,
+      error: null,
+      model: 'sora-2-pro',
+      progress: 100,
+      remixed_from_video_id: null,
+      seconds: '8',
+      size: '1792x1024',
+      status: 'completed'
+    }
+  },
+  {
+    id: '5',
+    type: 'openai',
+    status: 'failed',
+    error: { code: '400', message: 'Video generation failed' },
+    prompt: 'Mountain landscape with snow peaks and forest',
+    providerId: 'openai',
+    name: 'video-005',
+    metadata: {
+      id: 'video-005',
+      object: 'video',
+      created_at: Math.floor(Date.now() / 1000),
+      completed_at: Math.floor(Date.now() / 1000),
+      expires_at: null,
+      error: { code: '400', message: 'Video generation failed' },
+      model: 'sora-2',
+      progress: 0,
+      remixed_from_video_id: null,
+      seconds: '4',
+      size: '1280x720',
+      status: 'failed'
+    }
+  },
+  {
+    id: '6',
+    type: 'openai',
+    status: 'completed',
+    thumbnail: 'https://picsum.photos/200/200?random=6',
+    prompt: 'City street at night with neon lights reflecting on wet pavement',
+    providerId: 'openai',
+    name: 'video-006',
+    metadata: {
+      id: 'video-006',
+      object: 'video',
+      created_at: Math.floor(Date.now() / 1000),
+      completed_at: Math.floor(Date.now() / 1000),
+      expires_at: null,
+      error: null,
+      model: 'sora-2-pro',
+      progress: 100,
+      remixed_from_video_id: null,
+      seconds: '12',
+      size: '1024x1792',
+      status: 'completed'
+    }
+  }
+]

src/renderer/src/config/tools.ts

@@ -1,5 +1,5 @@
+import { ChatCompletionTool } from '@cherrystudio/openai/resources'
 import { Model } from '@renderer/types'
-import { ChatCompletionTool } from 'openai/resources'
 
 import { WEB_SEARCH_PROMPT_FOR_ZHIPU } from './prompts'
 

src/renderer/src/hooks/video/useAddOpenAIVideo.ts

@@ -0,0 +1,65 @@
+import OpenAI from '@cherrystudio/openai'
+import { useCallback } from 'react'
+
+import { useVideos } from './useVideos'
+
+export const useAddOpenAIVideo = (providerId: string) => {
+  const { addVideo } = useVideos(providerId)
+
+  const addOpenAIVideo = useCallback(
+    (video: OpenAI.Videos.Video, prompt: string) => {
+      switch (video.status) {
+        case 'queued':
+          addVideo({
+            id: video.id,
+            name: video.id,
+            providerId,
+            status: video.status,
+            type: 'openai',
+            metadata: video,
+            prompt
+          })
+          break
+        case 'in_progress':
+          addVideo({
+            id: video.id,
+            name: video.id,
+            providerId,
+            status: 'in_progress',
+            type: 'openai',
+            progress: video.progress,
+            metadata: video,
+            prompt
+          })
+          break
+        case 'completed':
+          addVideo({
+            id: video.id,
+            name: video.id,
+            providerId,
+            status: 'completed',
+            type: 'openai',
+            metadata: video,
+            prompt,
+            thumbnail: null
+          })
+          break
+        case 'failed':
+          addVideo({
+            id: video.id,
+            name: video.id,
+            providerId,
+            status: 'failed',
+            type: 'openai',
+            error: video.error,
+            metadata: video,
+            prompt
+          })
+          break
+      }
+    },
+    [addVideo]
+  )
+
+  return addOpenAIVideo
+}

src/renderer/src/hooks/video/useOpenAIVideo.ts

@@ -0,0 +1,43 @@
+import { retrieveVideo } from '@renderer/services/ApiService'
+import { SystemProviderIds } from '@renderer/types'
+import useSWR, { SWRConfiguration, useSWRConfig } from 'swr'
+
+import { useProvider } from '../useProvider'
+import { useVideo } from './useVideo'
+
+export const useOpenAIVideo = (id: string) => {
+  const providerId = SystemProviderIds.openai
+  const { provider: openai } = useProvider(providerId)
+  const fetcher = async () => {
+    return retrieveVideo({
+      type: 'openai',
+      videoId: id,
+      provider: openai
+    })
+  }
+  const video = useVideo(providerId, id)
+  let options: SWRConfiguration = {}
+  switch (video?.status) {
+    case 'queued':
+    case 'in_progress':
+      options = {
+        refreshInterval: 3000
+      }
+      break
+    default:
+      options = {
+        revalidateOnFocus: false,
+        revalidateOnMount: true
+      }
+  }
+  const { data, isLoading, error } = useSWR(`video/openai/${id}`, fetcher, options)
+  const { mutate } = useSWRConfig()
+  const revalidate = () => mutate(`video/openai/${id}`)
+
+  return {
+    video: data,
+    isLoading,
+    error,
+    revalidate
+  }
+}

src/renderer/src/hooks/video/useVideo.ts

@@ -0,0 +1,7 @@
+import { useVideos } from './useVideos'
+
+export const useVideo = (providerId: string, id: string) => {
+  const { videos } = useVideos(providerId)
+  const video = videos.find((v) => v.id === id)
+  return video
+}

src/renderer/src/hooks/video/useVideos.ts

@@ -0,0 +1,150 @@
+import { loggerService } from '@logger'
+import { retrieveVideo, retrieveVideoContent } from '@renderer/services/ApiService'
+import { getProviderById } from '@renderer/services/ProviderService'
+import { useAppDispatch, useAppSelector } from '@renderer/store'
+import {
+  addVideoAction,
+  removeVideoAction,
+  setVideoAction,
+  setVideosAction,
+  updateVideoAction
+} from '@renderer/store/video'
+import { Video } from '@renderer/types/video'
+import { useCallback, useEffect, useRef } from 'react'
+import useSWR from 'swr'
+
+const logger = loggerService.withContext('useVideo')
+
+export const useVideos = (providerId: string) => {
+  const videos = useAppSelector((state) => state.video.videoMap[providerId])
+  const videosRef = useRef(videos)
+  const dispatch = useAppDispatch()
+
+  useEffect(() => {
+    videosRef.current = videos
+  }, [videos])
+
+  const addVideo = useCallback(
+    (video: Video) => {
+      if (videos && videos.every((v) => v.id !== video.id)) {
+        dispatch(addVideoAction({ providerId, video }))
+      }
+    },
+    [dispatch, providerId, videos]
+  )
+
+  const updateVideo = useCallback(
+    (update: Partial<Omit<Video, 'status'>> & { id: string }) => {
+      dispatch(updateVideoAction({ providerId, update }))
+    },
+    [dispatch, providerId]
+  )
+
+  const setVideo = useCallback(
+    (video: Video) => {
+      dispatch(setVideoAction({ providerId, video }))
+    },
+    [dispatch, providerId]
+  )
+
+  const setVideos = useCallback(
+    (newVideos: Video[]) => {
+      dispatch(setVideosAction({ providerId, videos: newVideos }))
+    },
+    [dispatch, providerId]
+  )
+
+  const removeVideo = useCallback(
+    (videoId: string) => {
+      dispatch(removeVideoAction({ providerId, videoId }))
+    },
+    [dispatch, providerId]
+  )
+
+  useEffect(() => {
+    if (!videos) {
+      setVideos([])
+    }
+  }, [setVideos, videos])
+
+  // update videos from api
+  // NOTE: This provider should support openai videos endpoint. No runtime check here.
+  const provider = getProviderById(providerId)
+  const fetcher = async () => {
+    if (!videos || !provider) return []
+    const openaiVideos = videos.filter((v) => v.type === 'openai')
+    const jobs = openaiVideos.map((v) => retrieveVideo({ type: 'openai', videoId: v.id, provider }))
+    const result = await Promise.allSettled(jobs)
+    return result.filter((p) => p.status === 'fulfilled').map((p) => p.value)
+  }
+  const { data, error } = useSWR('video/openai/videos', fetcher, { refreshInterval: 3000 })
+  useEffect(() => {
+    if (error) {
+      logger.error('Failed to fetch video status updates', error)
+      return
+    }
+    if (!provider) {
+      logger.warn(`Provider ${providerId} not found.`)
+      return
+    }
+    const videos = videosRef.current
+
+    if (!data || !videos) return
+    data.forEach((v) => {
+      const retrievedVideo = v.video
+      const storeVideo = videos.find((v) => v.id === retrievedVideo.id)
+      if (!storeVideo) {
+        logger.warn(`Try to update video ${retrievedVideo.id}, but it's not in the store.`)
+        return
+      }
+      switch (retrievedVideo.status) {
+        case 'in_progress':
+          if (storeVideo.status === 'queued' || storeVideo.status === 'in_progress') {
+            setVideo({
+              ...storeVideo,
+              status: 'in_progress',
+              progress: retrievedVideo.progress,
+              metadata: retrievedVideo
+            })
+          }
+          break
+        case 'completed':
+          // Only update it when in_progress/queued -> completed
+          if (storeVideo.status === 'in_progress' || storeVideo.status === 'queued') {
+            setVideo({ ...storeVideo, status: 'completed', thumbnail: null, metadata: retrievedVideo })
+            // try to request thumbnail here.
+            retrieveVideoContent({
+              type: 'openai',
+              provider,
+              videoId: retrievedVideo.id,
+              query: { variant: 'thumbnail' }
+            })
+              .then((v) => {
+                // TODO: this is a iamge/webp type response. save it somewhere.
+                logger.debug('thumbnail resposne', v.response)
+              })
+              .catch((e) => {
+                logger.error(`Failed to get thumbnail for video ${retrievedVideo.id}`, e as Error)
+              })
+          }
+          break
+        case 'failed':
+          setVideo({
+            ...storeVideo,
+            status: 'failed',
+            error: retrievedVideo.error,
+            metadata: retrievedVideo
+          })
+      }
+    })
+  }, [data, error, provider, providerId, setVideo])
+
+  return {
+    videos: videos ?? [],
+    addVideo,
+    updateVideo,
+    setVideos,
+    setVideo,
+    removeVideo
+  }
+}

src/renderer/src/i18n/label.ts

@@ -146,7 +146,8 @@ const titleKeyMap = {
   notes: 'title.notes',
   paintings: 'title.paintings',
   settings: 'title.settings',
-  translate: 'title.translate'
+  translate: 'title.translate',
+  video: 'title.video'
 } as const
 
 export const getTitleLabel = (key: string): string => {
@@ -172,7 +173,8 @@ const sidebarIconKeyMap = {
   knowledge: 'knowledge.title',
   files: 'files.title',
   code_tools: 'code.title',
-  notes: 'notes.title'
+  notes: 'notes.title',
+  video: 'video.title'
 } as const
 
 export const getSidebarIconLabel = (key: string): string => {

src/renderer/src/i18n/locales/en-us.json

@@ -1014,10 +1014,12 @@
     "prompt": "Prompt",
     "provider": "Provider",
     "reasoning_content": "Deep reasoning",
+    "redownload": "Redownload",
     "refresh": "Refresh",
     "regenerate": "Regenerate",
     "rename": "Rename",
     "reset": "Reset",
+    "retry": "Retry",
     "save": "Save",
     "saved": "Saved",
     "search": "Search",
@@ -1025,6 +1027,7 @@
     "selected": "Selected",
     "selectedItems": "Selected {{count}} items",
     "selectedMessages": "Selected {{count}} messages",
+    "send": "Send",
     "settings": "Settings",
     "sort": {
       "pinyin": {
@@ -1191,7 +1194,8 @@
     "size": "Size",
     "text": "Text",
     "title": "Files",
-    "type": "Type"
+    "type": "Type",
+    "video": "Video"
   },
   "gpustack": {
     "keep_alive_time": {
@@ -4486,7 +4490,8 @@
     "paintings": "Paintings",
     "settings": "Settings",
     "store": "Assistant Library",
-    "translate": "Translate"
+    "translate": "Translate",
+    "video": "Video"
   },
   "trace": {
     "backList": "Back To List",
@@ -4644,6 +4649,44 @@
     "saveDataError": "Failed to save data, please try again.",
     "title": "Update"
   },
+  "video": {
+    "error": {
+      "create": "Failed to create video",
+      "download": "Failed to download video.",
+      "invalid": "Invalid video",
+      "load": {
+        "message": "Failed to load the video",
+        "reason": "The file may be corrupted or has been deleted externally."
+      }
+    },
+    "expired": "Expired",
+    "input_reference": {
+      "add": {
+        "error": {
+          "format": "Not a image",
+          "size": "This image is too large. It should be under 5MB."
+        },
+        "tooltip": "Add image reference"
+      }
+    },
+    "prompt": {
+      "placeholder": "describes the video to generate"
+    },
+    "seconds": "Seconds",
+    "size": "Size",
+    "status": {
+      "completed": "Generation Completed",
+      "downloading": "Downloading",
+      "failed": "Generation Failed",
+      "in_progress": "Generating",
+      "queued": "Queued"
+    },
+    "thumbnail": {
+      "placeholder": "No thumbnail"
+    },
+    "title": "Video",
+    "undefined": "No available video"
+  },
   "warning": {
     "missing_provider": "The supplier does not exist; reverted to the default supplier {{provider}}. This may cause issues."
   },

src/renderer/src/i18n/locales/zh-cn.json

@@ -1014,10 +1014,12 @@
     "prompt": "提示词",
     "provider": "提供商",
     "reasoning_content": "已深度思考",
+    "redownload": "重新下载",
     "refresh": "刷新",
     "regenerate": "重新生成",
     "rename": "重命名",
     "reset": "重置",
+    "retry": "重试",
     "save": "保存",
     "saved": "已保存",
     "search": "搜索",
@@ -1025,6 +1027,7 @@
     "selected": "已选择",
     "selectedItems": "已选择 {{count}} 项",
     "selectedMessages": "选中 {{count}} 条消息",
+    "send": "发送",
     "settings": "设置",
     "sort": {
       "pinyin": {
@@ -4486,7 +4489,8 @@
     "paintings": "绘画",
     "settings": "设置",
     "store": "助手库",
-    "translate": "翻译"
+    "translate": "翻译",
+    "video": "视频"
   },
   "trace": {
     "backList": "返回列表",
@@ -4644,6 +4648,42 @@
     "saveDataError": "保存数据失败，请重试",
     "title": "更新提示"
   },
+  "video": {
+    "error": {
+      "create": "创建视频失败",
+      "invalid": "无效的视频",
+      "load": {
+        "message": "加载视频失败",
+        "reason": "文件可能已损坏或已被外部删除。"
+      }
+    },
+    "input_reference": {
+      "add": {
+        "error": {
+          "format": "需要上传图片格式的文件",
+          "size": "图片过大，应小于 5MB"
+        },
+        "tooltip": "添加图像参考"
+      }
+    },
+    "prompt": {
+      "placeholder": "描述要生成的视频"
+    },
+    "seconds": "秒数",
+    "size": "尺寸",
+    "status": {
+      "completed": "生成完成",
+      "downloading": "下载中",
+      "failed": "生成失败",
+      "in_progress": "生成中",
+      "queued": "排队中"
+    },
+    "thumbnail": {
+      "placeholder": "无缩略图"
+    },
+    "title": "视频",
+    "undefined": "无可用视频"
+  },
   "warning": {
     "missing_provider": "供应商不存在，已回退到默认供应商 {{provider}}。这可能导致问题。"
   },

src/renderer/src/i18n/locales/zh-tw.json

@@ -1014,10 +1014,12 @@
     "prompt": "提示詞",
     "provider": "供應商",
     "reasoning_content": "已深度思考",
+    "redownload": "[to be translated]:Redownload",
     "refresh": "重新整理",
     "regenerate": "重新生成",
     "rename": "重新命名",
     "reset": "重設",
+    "retry": "[to be translated]:Retry",
     "save": "儲存",
     "saved": "已儲存",
     "search": "搜尋",
@@ -1025,6 +1027,7 @@
     "selected": "已選擇",
     "selectedItems": "已選擇 {{count}} 項",
     "selectedMessages": "選中 {{count}} 條訊息",
+    "send": "[to be translated]:Send",
     "settings": "設定",
     "sort": {
       "pinyin": {
@@ -4486,7 +4489,8 @@
     "paintings": "繪畫",
     "settings": "設定",
     "store": "助手庫",
-    "translate": "翻譯"
+    "translate": "翻譯",
+    "video": "[to be translated]:Video"
   },
   "trace": {
     "backList": "返回清單",
@@ -4644,6 +4648,42 @@
     "saveDataError": "保存數據失敗，請重試",
     "title": "更新提示"
   },
+  "video": {
+    "error": {
+      "create": "[to be translated]:Failed to create video",
+      "invalid": "[to be translated]:Invalid video",
+      "load": {
+        "message": "[to be translated]:Failed to load the video",
+        "reason": "[to be translated]:The file may be corrupted or has been deleted externally."
+      }
+    },
+    "input_reference": {
+      "add": {
+        "error": {
+          "format": "[to be translated]:Not a image",
+          "size": "[to be translated]:This image is too large. It should be under 5MB."
+        },
+        "tooltip": "[to be translated]:Add image reference"
+      }
+    },
+    "prompt": {
+      "placeholder": "[to be translated]:describes the video to generate"
+    },
+    "seconds": "[to be translated]:Seconds",
+    "size": "[to be translated]:Size",
+    "status": {
+      "completed": "[to be translated]:Generation Completed",
+      "downloading": "[to be translated]:Downloading",
+      "failed": "[to be translated]:Failed to generate video",
+      "in_progress": "[to be translated]:Generating",
+      "queued": "[to be translated]:Queued"
+    },
+    "thumbnail": {
+      "placeholder": "[to be translated]:No thumbnail"
+    },
+    "title": "[to be translated]:Video",
+    "undefined": "[to be translated]:No available video"
+  },
   "warning": {
     "missing_provider": "供應商不存在，已回退到預設供應商 {{provider}}。這可能導致問題。"
   },

src/renderer/src/i18n/translate/el-gr.json

@@ -1014,10 +1014,12 @@
     "prompt": "Ενδεικτικός ρήματος",
     "provider": "Παρέχων",
     "reasoning_content": "Έχει σκεφτεί πολύ καλά",
+    "redownload": "[to be translated]:Redownload",
     "refresh": "Ανανέωση",
     "regenerate": "Ξαναπαραγωγή",
     "rename": "Μετονομασία",
     "reset": "Επαναφορά",
+    "retry": "[to be translated]:Retry",
     "save": "Αποθήκευση",
     "saved": "Αποθηκεύτηκε",
     "search": "Αναζήτηση",
@@ -1025,6 +1027,7 @@
     "selected": "Επιλεγμένο",
     "selectedItems": "Επιλέχθηκαν {{count}} αντικείμενα",
     "selectedMessages": "Επιλέχθηκαν {{count}} μηνύματα",
+    "send": "[to be translated]:Send",
     "settings": "Ρυθμίσεις",
     "sort": {
       "pinyin": {
@@ -4486,7 +4489,8 @@
     "paintings": "Ζωγραφική",
     "settings": "Ρυθμίσεις",
     "store": "Βιβλιοθήκη βοηθών",
-    "translate": "Μετάφραση"
+    "translate": "Μετάφραση",
+    "video": "[to be translated]:Video"
   },
   "trace": {
     "backList": "Επιστροφή στη λίστα",
@@ -4644,6 +4648,42 @@
     "saveDataError": "Η αποθήκευση των δεδομένων απέτυχε, δοκιμάστε ξανά",
     "title": "Ενημέρωση"
   },
+  "video": {
+    "error": {
+      "create": "[to be translated]:Failed to create video",
+      "invalid": "[to be translated]:Invalid video",
+      "load": {
+        "message": "[to be translated]:Failed to load the video",
+        "reason": "[to be translated]:The file may be corrupted or has been deleted externally."
+      }
+    },
+    "input_reference": {
+      "add": {
+        "error": {
+          "format": "[to be translated]:Not a image",
+          "size": "[to be translated]:This image is too large. It should be under 5MB."
+        },
+        "tooltip": "[to be translated]:Add image reference"
+      }
+    },
+    "prompt": {
+      "placeholder": "[to be translated]:describes the video to generate"
+    },
+    "seconds": "[to be translated]:Seconds",
+    "size": "[to be translated]:Size",
+    "status": {
+      "completed": "[to be translated]:Generation Completed",
+      "downloading": "[to be translated]:Downloading",
+      "failed": "[to be translated]:Failed to generate video",
+      "in_progress": "[to be translated]:Generating",
+      "queued": "[to be translated]:Queued"
+    },
+    "thumbnail": {
+      "placeholder": "[to be translated]:No thumbnail"
+    },
+    "title": "[to be translated]:Video",
+    "undefined": "[to be translated]:No available video"
+  },
   "warning": {
     "missing_provider": "Ο προμηθευτής δεν υπάρχει, έγινε επαναφορά στον προεπιλεγμένο προμηθευτή {{provider}}. Αυτό μπορεί να προκαλέσει προβλήματα."
   },

src/renderer/src/i18n/translate/es-es.json

@@ -1014,10 +1014,12 @@
     "prompt": "Prompt",
     "provider": "Proveedor",
     "reasoning_content": "Pensamiento profundo",
+    "redownload": "[to be translated]:Redownload",
     "refresh": "Actualizar",
     "regenerate": "Regenerar",
     "rename": "Renombrar",
     "reset": "Restablecer",
+    "retry": "[to be translated]:Retry",
     "save": "Guardar",
     "saved": "Guardado",
     "search": "Buscar",
@@ -1025,6 +1027,7 @@
     "selected": "Seleccionado",
     "selectedItems": "{{count}} elementos seleccionados",
     "selectedMessages": "{{count}} mensajes seleccionados",
+    "send": "[to be translated]:Send",
     "settings": "Configuración",
     "sort": {
       "pinyin": {
@@ -4486,7 +4489,8 @@
     "paintings": "Pinturas",
     "settings": "Configuración",
     "store": "Biblioteca de asistentes",
-    "translate": "Traducir"
+    "translate": "Traducir",
+    "video": "[to be translated]:Video"
   },
   "trace": {
     "backList": "Volver a la lista",
@@ -4644,6 +4648,42 @@
     "saveDataError": "Error al guardar los datos, inténtalo de nuevo",
     "title": "Actualización"
   },
+  "video": {
+    "error": {
+      "create": "[to be translated]:Failed to create video",
+      "invalid": "[to be translated]:Invalid video",
+      "load": {
+        "message": "[to be translated]:Failed to load the video",
+        "reason": "[to be translated]:The file may be corrupted or has been deleted externally."
+      }
+    },
+    "input_reference": {
+      "add": {
+        "error": {
+          "format": "[to be translated]:Not a image",
+          "size": "[to be translated]:This image is too large. It should be under 5MB."
+        },
+        "tooltip": "[to be translated]:Add image reference"
+      }
+    },
+    "prompt": {
+      "placeholder": "[to be translated]:describes the video to generate"
+    },
+    "seconds": "[to be translated]:Seconds",
+    "size": "[to be translated]:Size",
+    "status": {
+      "completed": "[to be translated]:Generation Completed",
+      "downloading": "[to be translated]:Downloading",
+      "failed": "[to be translated]:Failed to generate video",
+      "in_progress": "[to be translated]:Generating",
+      "queued": "[to be translated]:Queued"
+    },
+    "thumbnail": {
+      "placeholder": "[to be translated]:No thumbnail"
+    },
+    "title": "[to be translated]:Video",
+    "undefined": "[to be translated]:No available video"
+  },
   "warning": {
     "missing_provider": "El proveedor no existe, se ha revertido al proveedor predeterminado {{provider}}. Esto podría causar problemas."
   },

src/renderer/src/i18n/translate/fr-fr.json

@@ -1014,10 +1014,12 @@
     "prompt": "Prompt",
     "provider": "Fournisseur",
     "reasoning_content": "Réflexion approfondie",
+    "redownload": "[to be translated]:Redownload",
     "refresh": "Actualiser",
     "regenerate": "Regénérer",
     "rename": "Renommer",
     "reset": "Réinitialiser",
+    "retry": "[to be translated]:Retry",
     "save": "Enregistrer",
     "saved": "enregistré",
     "search": "Rechercher",
@@ -1025,6 +1027,7 @@
     "selected": "Sélectionné",
     "selectedItems": "{{count}} éléments sélectionnés",
     "selectedMessages": "{{count}} messages sélectionnés",
+    "send": "[to be translated]:Send",
     "settings": "Paramètres",
     "sort": {
       "pinyin": {
@@ -4486,7 +4489,8 @@
     "paintings": "Peintures",
     "settings": "Paramètres",
     "store": "Bibliothèque d'assistants",
-    "translate": "Traduire"
+    "translate": "Traduire",
+    "video": "[to be translated]:Video"
   },
   "trace": {
     "backList": "Retour à la liste",
@@ -4644,6 +4648,42 @@
     "saveDataError": "Échec de la sauvegarde des données, veuillez réessayer",
     "title": "Mise à jour"
   },
+  "video": {
+    "error": {
+      "create": "[to be translated]:Failed to create video",
+      "invalid": "[to be translated]:Invalid video",
+      "load": {
+        "message": "[to be translated]:Failed to load the video",
+        "reason": "[to be translated]:The file may be corrupted or has been deleted externally."
+      }
+    },
+    "input_reference": {
+      "add": {
+        "error": {
+          "format": "[to be translated]:Not a image",
+          "size": "[to be translated]:This image is too large. It should be under 5MB."
+        },
+        "tooltip": "[to be translated]:Add image reference"
+      }
+    },
+    "prompt": {
+      "placeholder": "[to be translated]:describes the video to generate"
+    },
+    "seconds": "[to be translated]:Seconds",
+    "size": "[to be translated]:Size",
+    "status": {
+      "completed": "[to be translated]:Generation Completed",
+      "downloading": "[to be translated]:Downloading",
+      "failed": "[to be translated]:Failed to generate video",
+      "in_progress": "[to be translated]:Generating",
+      "queued": "[to be translated]:Queued"
+    },
+    "thumbnail": {
+      "placeholder": "[to be translated]:No thumbnail"
+    },
+    "title": "[to be translated]:Video",
+    "undefined": "[to be translated]:No available video"
+  },
   "warning": {
     "missing_provider": "Le fournisseur n’existe pas, retour au fournisseur par défaut {{provider}}. Cela peut entraîner des problèmes."
   },

src/renderer/src/i18n/translate/ja-jp.json

@@ -1014,10 +1014,12 @@
     "prompt": "プロンプト",
     "provider": "プロバイダー",
     "reasoning_content": "深く考察済み",
+    "redownload": "[to be translated]:Redownload",
     "refresh": "更新",
     "regenerate": "再生成",
     "rename": "名前を変更",
     "reset": "リセット",
+    "retry": "[to be translated]:Retry",
     "save": "保存",
     "saved": "保存されました",
     "search": "検索",
@@ -1025,6 +1027,7 @@
     "selected": "選択済み",
     "selectedItems": "{{count}}件の項目を選択しました",
     "selectedMessages": "{{count}}件のメッセージを選択しました",
+    "send": "[to be translated]:Send",
     "settings": "設定",
     "sort": {
       "pinyin": {
@@ -4486,7 +4489,8 @@
     "paintings": "ペインティング",
     "settings": "設定",
     "store": "アシスタントライブラリ",
-    "translate": "翻訳"
+    "translate": "翻訳",
+    "video": "[to be translated]:Video"
   },
   "trace": {
     "backList": "リストに戻る",
@@ -4644,6 +4648,42 @@
     "saveDataError": "データの保存に失敗しました。もう一度お試しください。",
     "title": "更新"
   },
+  "video": {
+    "error": {
+      "create": "[to be translated]:Failed to create video",
+      "invalid": "[to be translated]:Invalid video",
+      "load": {
+        "message": "[to be translated]:Failed to load the video",
+        "reason": "[to be translated]:The file may be corrupted or has been deleted externally."
+      }
+    },
+    "input_reference": {
+      "add": {
+        "error": {
+          "format": "[to be translated]:Not a image",
+          "size": "[to be translated]:This image is too large. It should be under 5MB."
+        },
+        "tooltip": "[to be translated]:Add image reference"
+      }
+    },
+    "prompt": {
+      "placeholder": "[to be translated]:describes the video to generate"
+    },
+    "seconds": "[to be translated]:Seconds",
+    "size": "[to be translated]:Size",
+    "status": {
+      "completed": "[to be translated]:Generation Completed",
+      "downloading": "[to be translated]:Downloading",
+      "failed": "[to be translated]:Failed to generate video",
+      "in_progress": "[to be translated]:Generating",
+      "queued": "[to be translated]:Queued"
+    },
+    "thumbnail": {
+      "placeholder": "[to be translated]:No thumbnail"
+    },
+    "title": "[to be translated]:Video",
+    "undefined": "[to be translated]:No available video"
+  },
   "warning": {
     "missing_provider": "サプライヤーが存在しないため、デフォルトのサプライヤー {{provider}} にロールバックされました。これにより問題が発生する可能性があります。"
   },

src/renderer/src/i18n/translate/pt-pt.json

@@ -1014,10 +1014,12 @@
     "prompt": "Prompt",
     "provider": "Fornecedor",
     "reasoning_content": "Pensamento profundo concluído",
+    "redownload": "[to be translated]:Redownload",
     "refresh": "Atualizar",
     "regenerate": "Regenerar",
     "rename": "Renomear",
     "reset": "Redefinir",
+    "retry": "[to be translated]:Retry",
     "save": "Salvar",
     "saved": "Guardado",
     "search": "Pesquisar",
@@ -1025,6 +1027,7 @@
     "selected": "Selecionado",
     "selectedItems": "{{count}} itens selecionados",
     "selectedMessages": "{{count}} mensagens selecionadas",
+    "send": "[to be translated]:Send",
     "settings": "Configurações",
     "sort": {
       "pinyin": {
@@ -4486,7 +4489,8 @@
     "paintings": "Pinturas",
     "settings": "Configurações",
     "store": "Biblioteca de assistentes",
-    "translate": "Traduzir"
+    "translate": "Traduzir",
+    "video": "[to be translated]:Video"
   },
   "trace": {
     "backList": "Voltar à lista",
@@ -4644,6 +4648,42 @@
     "saveDataError": "Falha ao salvar os dados, tente novamente",
     "title": "Atualização"
   },
+  "video": {
+    "error": {
+      "create": "[to be translated]:Failed to create video",
+      "invalid": "[to be translated]:Invalid video",
+      "load": {
+        "message": "[to be translated]:Failed to load the video",
+        "reason": "[to be translated]:The file may be corrupted or has been deleted externally."
+      }
+    },
+    "input_reference": {
+      "add": {
+        "error": {
+          "format": "[to be translated]:Not a image",
+          "size": "[to be translated]:This image is too large. It should be under 5MB."
+        },
+        "tooltip": "[to be translated]:Add image reference"
+      }
+    },
+    "prompt": {
+      "placeholder": "[to be translated]:describes the video to generate"
+    },
+    "seconds": "[to be translated]:Seconds",
+    "size": "[to be translated]:Size",
+    "status": {
+      "completed": "[to be translated]:Generation Completed",
+      "downloading": "[to be translated]:Downloading",
+      "failed": "[to be translated]:Failed to generate video",
+      "in_progress": "[to be translated]:Generating",
+      "queued": "[to be translated]:Queued"
+    },
+    "thumbnail": {
+      "placeholder": "[to be translated]:No thumbnail"
+    },
+    "title": "[to be translated]:Video",
+    "undefined": "[to be translated]:No available video"
+  },
   "warning": {
     "missing_provider": "O fornecedor não existe; foi revertido para o fornecedor predefinido {{provider}}. Isto pode causar problemas."
   },

src/renderer/src/i18n/translate/ru-ru.json

@@ -1014,10 +1014,12 @@
     "prompt": "Промпт",
     "provider": "Провайдер",
     "reasoning_content": "Глубокий анализ",
+    "redownload": "[to be translated]:Redownload",
     "refresh": "Обновить",
     "regenerate": "Пересоздать",
     "rename": "Переименовать",
     "reset": "Сбросить",
+    "retry": "[to be translated]:Retry",
     "save": "Сохранить",
     "saved": "Сохранено",
     "search": "Поиск",
@@ -1025,6 +1027,7 @@
     "selected": "Выбрано",
     "selectedItems": "Выбрано {{count}} элементов",
     "selectedMessages": "Выбрано {{count}} сообщений",
+    "send": "[to be translated]:Send",
     "settings": "Настройки",
     "sort": {
       "pinyin": {
@@ -4486,7 +4489,8 @@
     "paintings": "Рисунки",
     "settings": "Настройки",
     "store": "Библиотека помощников",
-    "translate": "Перевод"
+    "translate": "Перевод",
+    "video": "[to be translated]:Video"
   },
   "trace": {
     "backList": "Вернуться к списку",
@@ -4644,6 +4648,42 @@
     "saveDataError": "Ошибка сохранения данных, повторите попытку",
     "title": "Обновление"
   },
+  "video": {
+    "error": {
+      "create": "[to be translated]:Failed to create video",
+      "invalid": "[to be translated]:Invalid video",
+      "load": {
+        "message": "[to be translated]:Failed to load the video",
+        "reason": "[to be translated]:The file may be corrupted or has been deleted externally."
+      }
+    },
+    "input_reference": {
+      "add": {
+        "error": {
+          "format": "[to be translated]:Not a image",
+          "size": "[to be translated]:This image is too large. It should be under 5MB."
+        },
+        "tooltip": "[to be translated]:Add image reference"
+      }
+    },
+    "prompt": {
+      "placeholder": "[to be translated]:describes the video to generate"
+    },
+    "seconds": "[to be translated]:Seconds",
+    "size": "[to be translated]:Size",
+    "status": {
+      "completed": "[to be translated]:Generation Completed",
+      "downloading": "[to be translated]:Downloading",
+      "failed": "[to be translated]:Failed to generate video",
+      "in_progress": "[to be translated]:Generating",
+      "queued": "[to be translated]:Queued"
+    },
+    "thumbnail": {
+      "placeholder": "[to be translated]:No thumbnail"
+    },
+    "title": "[to be translated]:Video",
+    "undefined": "[to be translated]:No available video"
+  },
   "warning": {
     "missing_provider": "Поставщик не существует, возвращение к поставщику по умолчанию {{provider}}. Это может привести к проблемам."
   },

src/renderer/src/pages/files/FilesPage.tsx

@@ -19,7 +19,8 @@ import {
   File as FileIcon,
   FileImage,
   FileText,
-  FileType as FileTypeIcon
+  FileType as FileTypeIcon,
+  FileVideo
 } from 'lucide-react'
 import { FC, useEffect, useState } from 'react'
 import { useTranslation } from 'react-i18next'
@@ -138,6 +139,7 @@ const FilesPage: FC = () => {
     { key: FileTypes.DOCUMENT, label: t('files.document'), icon: <FileIcon size={16} /> },
     { key: FileTypes.IMAGE, label: t('files.image'), icon: <FileImage size={16} /> },
     { key: FileTypes.TEXT, label: t('files.text'), icon: <FileTypeIcon size={16} /> },
+    { key: FileTypes.VIDEO, label: t('files.video'), icon: <FileVideo size={16} /> },
     { key: 'all', label: t('files.all'), icon: <FileText size={16} /> }
   ]
 

src/renderer/src/pages/launchpad/LaunchpadPage.tsx

@@ -2,7 +2,7 @@ import App from '@renderer/components/MinApp/MinApp'
 import { useMinapps } from '@renderer/hooks/useMinapps'
 import { useRuntime } from '@renderer/hooks/useRuntime'
 import { useSettings } from '@renderer/hooks/useSettings'
-import { Code, FileSearch, Folder, Languages, LayoutGrid, NotepadText, Palette, Sparkle } from 'lucide-react'
+import { Code, FileSearch, Folder, Languages, LayoutGrid, NotepadText, Palette, Sparkle, Video } from 'lucide-react'
 import { FC, useMemo } from 'react'
 import { useTranslation } from 'react-i18next'
 import { useNavigate } from 'react-router-dom'
@@ -63,6 +63,12 @@ const LaunchpadPage: FC = () => {
       text: t('title.notes'),
       path: '/notes',
       bgColor: 'linear-gradient(135deg, #F97316, #FB923C)' // 笔记：橙色，代表活力和清晰思路
+    },
+    {
+      icon: <Video size={32} className="icon" />,
+      text: t('title.video'),
+      path: '/video',
+      bgColor: 'linear-gradient(135deg, #7C3AED, #A78BFA)' // Video Generation: deep purple, representing creativity and dynamic media
     }
   ]
 

src/renderer/src/pages/settings/DisplaySettings/SidebarIconsManager.tsx

@@ -21,7 +21,8 @@ import {
   MessageSquareQuote,
   NotepadText,
   Palette,
-  Sparkle
+  Sparkle,
+  Video
 } from 'lucide-react'
 import { FC, useCallback, useMemo } from 'react'
 import { useTranslation } from 'react-i18next'
@@ -127,7 +128,8 @@ const SidebarIconsManager: FC<SidebarIconsManagerProps> = ({
       knowledge: <FileSearch size={16} />,
       files: <Folder size={16} />,
       notes: <NotepadText size={16} />,
-      code_tools: <Code size={16} />
+      code_tools: <Code size={16} />,
+      video: <Video size={16} />
     }),
     []
   )

src/renderer/src/pages/video/VideoList.tsx

@@ -0,0 +1,133 @@
+import { cn, Progress, Spinner } from '@heroui/react'
+import { Video } from '@renderer/types'
+import { CheckCircleIcon, CircleXIcon, ClockIcon, DownloadIcon, PlusIcon } from 'lucide-react'
+import { useTranslation } from 'react-i18next'
+
+export type VideoListProps = {
+  videos: Video[]
+  activeVideoId?: string
+  setActiveVideoId: (id: string | undefined) => void
+}
+
+export const VideoList = ({ videos, activeVideoId, setActiveVideoId }: VideoListProps) => {
+  return (
+    <div className="w-40 space-y-3 overflow-auto p-2">
+      <div
+        className="group relative flex aspect-square cursor-pointer items-center justify-center overflow-hidden rounded-xl border-2 transition-all hover:scale-105 hover:shadow-lg"
+        onClick={() => setActiveVideoId(undefined)}>
+        <PlusIcon size={24} />
+      </div>
+      {/* {mockVideos.map((video) => ( */}
+      {videos.map((video) => (
+        <VideoListItem
+          key={video.id}
+          video={video}
+          isActive={activeVideoId === video.id}
+          onClick={() => setActiveVideoId(video.id)}
+        />
+      ))}
+    </div>
+  )
+}
+
+const VideoListItem = ({ video, isActive, onClick }: { video: Video; isActive: boolean; onClick: () => void }) => {
+  const { t } = useTranslation()
+
+  const getStatusIcon = () => {
+    switch (video.status) {
+      case 'queued':
+        return <ClockIcon size={20} className="text-default-500" />
+      case 'in_progress':
+        return <Spinner size="sm" color="primary" />
+      case 'completed':
+        return <CheckCircleIcon size={20} className="text-success" />
+      case 'downloading':
+        return <DownloadIcon size={20} className="text-primary" />
+      case 'downloaded':
+        return null // No indicator for downloaded state
+      case 'failed':
+        return <CircleXIcon size={20} className="text-danger" />
+      default:
+        return null
+    }
+  }
+
+  const getStatusColor = () => {
+    switch (video.status) {
+      case 'queued':
+        return 'bg-default-100'
+      case 'in_progress':
+        return 'bg-primary-50'
+      case 'completed':
+        return 'bg-success-50'
+      case 'downloading':
+        return 'bg-primary-50'
+      case 'downloaded':
+        return 'bg-success-50'
+      case 'failed':
+        return 'bg-danger-50'
+      default:
+        return 'bg-default-50'
+    }
+  }
+
+  const showProgress = video.status === 'in_progress' || video.status === 'downloading'
+  const showThumbnail = video.status === 'completed' || video.status === 'downloading' || video.status === 'downloaded'
+
+  return (
+    <div
+      className={cn(
+        `group relative aspect-square cursor-pointer overflow-hidden rounded-xl border-2 transition-all hover:scale-105 hover:shadow-lg ${getStatusColor()}`,
+        isActive ? 'border-primary' : undefined
+      )}
+      onClick={onClick}>
+      {/* Thumbnail placeholder */}
+      <div className="absolute inset-0 flex items-center justify-center bg-gradient-to-br from-default-100 to-default-200">
+        {showThumbnail ? (
+          <img src={video.thumbnail ?? ''} alt="Video thumbnail" className="h-full w-full object-cover" />
+        ) : (
+          <div className="flex flex-col items-center gap-2 text-default-400">
+            <div className="text-2xl">🎬</div>
+          </div>
+        )}
+      </div>
+
+      {/* Status overlay */}
+      <div className="absolute inset-0 bg-black/20 opacity-0 transition-opacity group-hover:opacity-100" />
+
+      {/* Status indicator */}
+      {getStatusIcon() && (
+        <div className="absolute top-2 right-2 flex items-center gap-1 rounded-full bg-white/90 px-2 py-1 backdrop-blur-sm">
+          {getStatusIcon()}
+          <span className="font-medium text-xs">{t(`video.status.${video.status}`)}</span>
+        </div>
+      )}
+
+      {/* Progress bar for in_progress and downloading states */}
+      {showProgress && (
+        <div className="absolute right-0 bottom-0 left-0 p-2">
+          <Progress
+            size="sm"
+            value={video.progress}
+            color={video.status === 'downloading' ? 'primary' : 'primary'}
+            className="w-full"
+            showValueLabel={false}
+          />
+        </div>
+      )}
+
+      {/* Video info overlay */}
+      <div className="absolute right-0 bottom-0 left-0 bg-gradient-to-t from-black/60 to-transparent p-3 pt-6 opacity-0 transition-opacity group-hover:opacity-100">
+        <div className="text-white">
+          <p className="truncate font-medium text-sm">{video.metadata.id}</p>
+          {video.prompt && <p className="mt-1 line-clamp-2 text-xs opacity-80">{video.prompt}</p>}
+        </div>
+      </div>
+
+      {/* Failed state overlay */}
+      {video.status === 'failed' && (
+        <div className="absolute inset-0 flex items-center justify-center bg-danger/10"></div>
+      )}
+    </div>
+  )
+}

src/renderer/src/pages/video/VideoPage.tsx

@@ -0,0 +1,80 @@
+// interface VideoPageProps {}
+
+import { Divider } from '@heroui/react'
+import { Navbar, NavbarCenter } from '@renderer/components/app/Navbar'
+import { useProvider } from '@renderer/hooks/useProvider'
+import { useVideos } from '@renderer/hooks/video/useVideos'
+import { SystemProviderIds } from '@renderer/types'
+import { CreateVideoParams } from '@renderer/types/video'
+import { deepUpdate } from '@renderer/utils/deepUpdate'
+import { isVideoModel } from '@renderer/utils/model/video'
+import { DeepPartial } from 'ai'
+import { useCallback, useMemo, useState } from 'react'
+import { useTranslation } from 'react-i18next'
+
+import { ModelSetting } from './settings/ModelSetting'
+import { OpenAIParamSettings } from './settings/OpenAIParamSettings'
+import { ProviderSetting } from './settings/ProviderSetting'
+import { SettingsGroup } from './settings/shared'
+import { VideoList } from './VideoList'
+import { VideoPanel } from './VideoPanel'
+
+export const VideoPage = () => {
+  const { t } = useTranslation()
+  const [providerId, setProviderId] = useState<string>(SystemProviderIds.openai)
+  const { provider } = useProvider(providerId)
+  const [params, setParams] = useState<CreateVideoParams>({
+    type: 'openai',
+    provider,
+    params: {
+      model: 'sora-2',
+      prompt: ''
+    },
+    options: {}
+  })
+  const [activeVideoId, setActiveVideoId] = useState<string>()
+
+  const updateParams = useCallback((update: DeepPartial<Omit<CreateVideoParams, 'type'>>) => {
+    setParams((prev) => deepUpdate<CreateVideoParams>(prev, update))
+  }, [])
+
+  const updateModelId = useCallback(
+    (id: string) => {
+      if (isVideoModel(id)) {
+        updateParams({ params: { model: id } })
+      }
+    },
+    [updateParams]
+  )
+
+  const { videos } = useVideos(providerId)
+  // const activeVideo = useMemo(() => mockVideos.find((v) => v.id === activeVideoId), [activeVideoId])
+  const activeVideo = useMemo(() => videos.find((v) => v.id === activeVideoId), [activeVideoId, videos])
+
+  return (
+    <div className="flex flex-1 flex-col">
+      <Navbar>
+        <NavbarCenter style={{ borderRight: 'none' }}>{t('video.title')}</NavbarCenter>
+      </Navbar>
+      <div id="content-container" className="flex max-h-full flex-1">
+        {/* Settings */}
+        <div className="flex w-70 flex-col p-2">
+          <SettingsGroup>
+            <ProviderSetting providerId={providerId} setProviderId={setProviderId} />
+            <ModelSetting
+              providerId={providerId}
+              modelId={params.params.model ?? 'sora-2'}
+              setModelId={updateModelId}
+            />
+          </SettingsGroup>
+          {provider.type === 'openai-response' && <OpenAIParamSettings params={params} updateParams={updateParams} />}
+        </div>
+        <Divider orientation="vertical" />
+        <VideoPanel provider={provider} params={params} updateParams={updateParams} video={activeVideo} />
+        <Divider orientation="vertical" />
+        {/* Video list */}
+        <VideoList videos={videos} activeVideoId={activeVideoId} setActiveVideoId={setActiveVideoId} />
+      </div>
+    </div>
+  )
+}

src/renderer/src/pages/video/VideoPanel.tsx

@@ -0,0 +1,287 @@
+import { Button, cn, Image, Skeleton, Textarea, Tooltip } from '@heroui/react'
+import { loggerService } from '@logger'
+import { useAddOpenAIVideo } from '@renderer/hooks/video/useAddOpenAIVideo'
+import { useVideos } from '@renderer/hooks/video/useVideos'
+import { createVideo, retrieveVideoContent } from '@renderer/services/ApiService'
+import FileManager from '@renderer/services/FileManager'
+import { FileTypes, Provider, VideoFileMetadata } from '@renderer/types'
+import { CreateVideoParams, Video } from '@renderer/types/video'
+import { getErrorMessage } from '@renderer/utils'
+import { MB } from '@shared/config/constant'
+import { DeepPartial } from 'ai'
+import dayjs from 'dayjs'
+import { isEmpty } from 'lodash'
+import { ArrowUp, CircleXIcon, ImageIcon } from 'lucide-react'
+import mime from 'mime-types'
+import { useCallback, useEffect, useMemo, useRef, useState } from 'react'
+import { useTranslation } from 'react-i18next'
+
+import { VideoViewer } from './VideoViewer'
+
+export type VideoPanelProps = {
+  provider: Provider
+  video?: Video
+  params: CreateVideoParams
+  updateParams: (upadte: DeepPartial<Omit<CreateVideoParams, 'type'>>) => void
+}
+
+const logger = loggerService.withContext('VideoPanel')
+
+export const VideoPanel = ({ provider, video, params, updateParams }: VideoPanelProps) => {
+  const { t } = useTranslation()
+  const addOpenAIVideo = useAddOpenAIVideo(provider.id)
+  const { setVideo } = useVideos(provider.id)
+
+  const [isProcessing, setIsProcessing] = useState(false)
+  const fileInputRef = useRef<HTMLInputElement>(null)
+  const inputReference = params.params.input_reference
+
+  const couldCreateVideo = useMemo(
+    () =>
+      !isProcessing &&
+      !isEmpty(params.params.prompt) &&
+      video?.status !== 'queued' &&
+      video?.status !== 'downloading' &&
+      video?.status !== 'in_progress',
+    [isProcessing, params.params.prompt, video?.status]
+  )
+
+  useEffect(() => {
+    if (video) {
+      updateParams({ params: { prompt: video.prompt } })
+    } else {
+      updateParams({ params: { prompt: '' } })
+    }
+  }, [updateParams, video])
+
+  const handleCreateVideo = useCallback(async () => {
+    if (!couldCreateVideo) return
+    setIsProcessing(true)
+    try {
+      if (video === undefined) {
+        const result = await createVideo(params)
+        const video = result.video
+        switch (result.type) {
+          case 'openai':
+            addOpenAIVideo(video, params.params.prompt)
+            break
+          default:
+            logger.error(`Invalid video type ${result.type}.`)
+        }
+      } else {
+        // TODO: remix video
+        window.toast.info('Remix video is not implemented.')
+      }
+    } catch (e) {
+      window.toast.error({ title: t('video.error.create'), description: getErrorMessage(e), timeout: 5000 })
+    } finally {
+      setIsProcessing(false)
+    }
+  }, [addOpenAIVideo, couldCreateVideo, params, t, video])
+
+  const handleRegenerateVideo = useCallback(() => {
+    window.toast.info('Not implemented')
+  }, [])
+
+  const handleDownloadVideo = useCallback(async () => {
+    if (!video) return
+    if (video.status !== 'completed' && video.status !== 'downloaded') return
+
+    const baseVideo: Video = {
+      ...video,
+      status: 'downloading',
+      progress: 0,
+      thumbnail: video.thumbnail
+    }
+    setVideo(baseVideo)
+
+    try {
+      const { response } = await retrieveVideoContent({ type: 'openai', videoId: video.id, provider })
+      if (!response.body) {
+        throw new Error('Video response body is empty')
+      }
+
+      const reader = response.body.getReader()
+      const contentLengthHeader = response.headers.get('content-length')
+      const totalSize = contentLengthHeader ? Number(contentLengthHeader) : undefined
+      const chunks: Uint8Array[] = []
+      let receivedLength = 0
+      let progressValue = 0
+
+      while (true) {
+        const { done, value } = await reader.read()
+        if (done) break
+        if (!value) continue
+
+        chunks.push(value)
+        receivedLength += value.length
+
+        if (totalSize && Number.isFinite(totalSize) && totalSize > 0) {
+          progressValue = Math.floor((receivedLength / totalSize) * 100)
+        } else {
+          progressValue = Math.min(progressValue + 1, 99)
+        }
+
+        setVideo({
+          ...baseVideo,
+          progress: Math.min(progressValue, 99)
+        })
+      }
+
+      const fileData = new Uint8Array(receivedLength)
+      let offset = 0
+      for (const chunk of chunks) {
+        fileData.set(chunk, offset)
+        offset += chunk.length
+      }
+
+      const contentType = response.headers.get('content-type') ?? 'video/mp4'
+      const normalizedContentType = contentType.split(';')[0]?.trim() || 'video/mp4'
+      const extension = (() => {
+        const ext = mime.extension(normalizedContentType)
+        return ext ? `.${ext}` : '.mp4'
+      })()
+
+      const fileName = `${video.id}${extension}`.toLowerCase()
+
+      const tempFilePath = await window.api.file.createTempFile(fileName)
+      await window.api.file.write(tempFilePath, fileData)
+
+      const tempFileMetadata = {
+        id: crypto.randomUUID(),
+        name: fileName,
+        origin_name: fileName,
+        path: tempFilePath,
+        size: receivedLength,
+        ext: extension,
+        type: FileTypes.VIDEO,
+        created_at: dayjs().toISOString(),
+        count: 1
+      } satisfies VideoFileMetadata
+
+      const uploadedFile = await FileManager.uploadFile(tempFileMetadata)
+
+      setVideo({
+        ...video,
+        status: 'downloaded',
+        thumbnail: video.thumbnail,
+        fileId: uploadedFile.id,
+        name: uploadedFile.origin_name
+      })
+    } catch (error) {
+      logger.error(`Failed to download video ${video.id}.`, error as Error)
+      window.toast.error(t('video.error.download'))
+      setVideo(video)
+    }
+  }, [provider, setVideo, t, video])
+
+  const handleUploadFile = useCallback(() => {
+    fileInputRef.current?.click()
+  }, [])
+
+  const setPrompt = useCallback((value: string) => updateParams({ params: { prompt: value } }), [updateParams])
+
+  const UploadImageReferenceButton = useCallback(() => {
+    const content = inputReference ? (
+      <div className="group">
+        <Image
+          className="aspect-square max-h-50 max-w-50 object-contain"
+          src={URL.createObjectURL(inputReference as File)}
+        />
+        <Button
+          variant="light"
+          color="danger"
+          className="absolute top-1 right-1 z-100 h-6 w-6 min-w-0 opacity-0 group-hover:opacity-100"
+          isIconOnly
+          startContent={<CircleXIcon size={16} className="text-danger" />}
+          onPress={() => updateParams({ params: { input_reference: undefined } })}
+        />
+      </div>
+    ) : (
+      t('video.input_reference.add.tooltip')
+    )
+    return (
+      <>
+        <Tooltip content={content} closeDelay={0}>
+          <Button
+            variant="light"
+            startContent={<ImageIcon size={16} className={cn(inputReference ? 'text-primary' : undefined)} />}
+            isIconOnly
+            className="h-6 w-6 min-w-0"
+            isDisabled={isProcessing}
+            onPress={handleUploadFile}
+          />
+        </Tooltip>
+      </>
+    )
+  }, [handleUploadFile, isProcessing, inputReference, t, updateParams])
+
+  return (
+    <div className="flex flex-1 flex-col p-2">
+      <div className="m-8 flex-1 overflow-hidden">
+        <Skeleton className="h-full w-full rounded-2xl" classNames={{ content: 'h-full w-full' }} isLoaded={true}>
+          {video && <VideoViewer video={video} onDownload={handleDownloadVideo} onRegenerate={handleRegenerateVideo} />}
+          {!video && <VideoViewer video={video} />}
+        </Skeleton>
+      </div>
+      <div className="relative">
+        <Textarea
+          label={t('common.prompt')}
+          placeholder={t('video.prompt.placeholder')}
+          value={params.params.prompt}
+          onValueChange={setPrompt}
+          isClearable
+          isDisabled={isProcessing}
+          classNames={{ inputWrapper: 'pb-8' }}
+          onKeyDown={(e: React.KeyboardEvent) => {
+            if (e.key === 'Enter') {
+              e.preventDefault()
+              handleCreateVideo()
+            }
+          }}
+        />
+        <div className="absolute bottom-0 flex w-full items-end justify-between p-2">
+          <div className="flex">
+            <UploadImageReferenceButton />
+            <input
+              ref={fileInputRef}
+              type="file"
+              hidden
+              onChange={(e) => {
+                const files = e.target.files
+                if (files && files.length > 0) {
+                  const file = files[0]
+                  if (!file.type.startsWith('image/')) {
+                    window.toast.error(t('video.input_reference.add.error.format'))
+                    return
+                  }
+                  const maxSize = 5 * MB
+                  if (file.size > maxSize) {
+                    window.toast.error(t('video.input_reference.add.error.size'))
+                    return
+                  }
+                  updateParams({ params: { input_reference: file } })
+                } else {
+                  updateParams({ params: { input_reference: undefined } })
+                }
+              }}
+            />
+          </div>
+
+          <Tooltip content={t('common.send')} closeDelay={0}>
+            <Button
+              color="primary"
+              radius="full"
+              isIconOnly
+              isDisabled={!couldCreateVideo}
+              isLoading={isProcessing}
+              className="h-6 w-6 min-w-0"
+              onPress={handleCreateVideo}>
+              <ArrowUp size={16} className="text-primary-foreground" />
+            </Button>
+          </Tooltip>
+        </div>
+      </div>
+    </div>
+  )
+}

src/renderer/src/pages/video/VideoViewer.tsx

@@ -0,0 +1,213 @@
+import {
+  Alert,
+  Button,
+  Modal,
+  ModalBody,
+  ModalContent,
+  ModalFooter,
+  Progress,
+  Skeleton,
+  Spinner,
+  useDisclosure
+} from '@heroui/react'
+import FileManager from '@renderer/services/FileManager'
+import { Video, VideoDownloaded, VideoFailed } from '@renderer/types/video'
+import dayjs from 'dayjs'
+import { CheckCircleIcon, CircleXIcon, Clock9Icon } from 'lucide-react'
+import { useEffect, useMemo, useState } from 'react'
+import { useTranslation } from 'react-i18next'
+import useSWRImmutable from 'swr/immutable'
+
+export type VideoViewerProps =
+  | {
+      video: undefined
+      onDownload?: never
+      onRegenerate?: never
+    }
+  | {
+      video: Video
+      onDownload: () => void
+      onRegenerate: () => void
+    }
+
+export const VideoViewer = ({ video, onDownload, onRegenerate }: VideoViewerProps) => {
+  const { t } = useTranslation()
+  const [loadSuccess, setLoadSuccess] = useState<boolean | undefined>(undefined)
+  useEffect(() => {
+    setLoadSuccess(undefined)
+  }, [video?.id])
+  return (
+    <>
+      <div className="flex h-full max-h-full w-full items-center justify-center rounded-2xl bg-foreground-200">
+        {video === undefined && t('video.undefined')}
+        {video && video.status === 'queued' && <QueuedVideo />}
+        {video && video.status === 'in_progress' && <InProgressVideo progress={video.progress} />}
+        {video && video.status === 'completed' && (
+          <CompletedVideo video={video} onDownload={onDownload} onRegenerate={onRegenerate} />
+        )}
+        {video && video.status === 'downloading' && <DownloadingVideo progress={video.progress} />}
+        {video && video.status === 'downloaded' && loadSuccess !== false && (
+          <VideoPlayer video={video} setLoadSuccess={setLoadSuccess} />
+        )}
+        {video && video.status === 'failed' && <FailedVideo error={video.error} />}
+        {video && video.status === 'downloaded' && loadSuccess === false && (
+          <LoadFailedVideo onRedownload={onDownload} />
+        )}
+      </div>
+    </>
+  )
+}
+
+const QueuedVideo = () => {
+  const { t } = useTranslation()
+  return (
+    <div className="flex h-full w-full flex-col items-center justify-center rounded-2xl">
+      <Spinner variant="dots" />
+      <span>{t('video.status.queued')}</span>
+    </div>
+  )
+}
+
+const InProgressVideo = ({ progress }: { progress: number }) => {
+  const { t } = useTranslation()
+  return (
+    <div className="flex h-full w-full flex-col items-center justify-center rounded-2xl">
+      <Progress
+        label={t('video.status.in_progress')}
+        aria-label={t('video.status.in_progress')}
+        className="max-w-md"
+        color="primary"
+        showValueLabel={true}
+        size="md"
+        value={progress}
+      />
+    </div>
+  )
+}
+
+const CompletedVideo = ({
+  video,
+  onDownload,
+  onRegenerate
+}: {
+  video: Video
+  onDownload: () => void
+  onRegenerate: () => void
+}) => {
+  const { t } = useTranslation()
+  const isExpired = video.metadata.expires_at !== null && video.metadata.expires_at < dayjs().unix()
+  if (isExpired) {
+    return (
+      <div className="flex h-full w-full flex-col items-center justify-center gap-2 rounded-2xl bg-warning-200">
+        <Clock9Icon size={64} className="text-warning" />
+        <span className="font-bold text-2xl">{t('video.expired')}</span>
+        <Button onPress={onRegenerate}>{t('common.regenerate')}</Button>
+      </div>
+    )
+  }
+  return (
+    <div className="flex h-full w-full flex-col items-center justify-center gap-2 rounded-2xl bg-success-200">
+      <CheckCircleIcon size={64} className="text-success" />
+      <span className="font-bold text-2xl">{t('video.status.completed')}</span>
+      <Button onPress={onDownload}>{t('common.download')}</Button>
+    </div>
+  )
+}
+
+const DownloadingVideo = ({ progress }: { progress?: number }) => {
+  const { t } = useTranslation()
+  return (
+    <div className="flex h-full w-full flex-col items-center justify-center rounded-2xl">
+      <Progress
+        label={t('video.status.downloading')}
+        aria-label={t('video.status.downloading')}
+        className="max-w-md"
+        color="primary"
+        showValueLabel={true}
+        size="md"
+        value={progress}
+      />
+    </div>
+  )
+}
+
+const FailedVideo = ({ error }: { error: VideoFailed['error'] }) => {
+  const { t } = useTranslation()
+  const { isOpen, onOpen, onClose } = useDisclosure()
+
+  const alert = useMemo(() => {
+    if (error === null) {
+      return <Alert color="danger" title={t('error.unknown')} />
+    } else {
+      return <Alert color="danger" title={error.code} description={error.message} />
+    }
+  }, [error, t])
+
+  return (
+    <div className="flex h-full w-full flex-col items-center justify-center rounded-2xl bg-danger-200">
+      <CircleXIcon size={64} className="fill-danger text-danger-200" />
+      <span className="font-bold text-2xl">{t('video.status.failed')}</span>
+      <div className="my-2 flex justify-between gap-2">
+        <Button onPress={onOpen}>{t('common.detail')}</Button>
+        <Modal isOpen={isOpen} onClose={onClose}>
+          <ModalBody>
+            <ModalContent>
+              <div className="p-4">{alert}</div>
+            </ModalContent>
+          </ModalBody>
+          <ModalFooter></ModalFooter>
+        </Modal>
+        <Button onPress={() => window.toast.info('Not implemented')}>{t('common.retry')}</Button>
+      </div>
+    </div>
+  )
+}
+
+const LoadFailedVideo = ({ onRedownload }: { onRedownload: () => void }) => {
+  const { t } = useTranslation()
+  return (
+    <div className="flex h-full w-full flex-col items-center justify-center rounded-2xl bg-danger-200">
+      <CircleXIcon size={64} className="fill-danger text-danger-200" />
+      <span className="font-bold text-2xl">{t('video.error.load.message')}</span>
+      <span>{t('video.error.load.reason')}</span>
+      <div className="my-2 flex justify-between gap-2">
+        <Button onPress={onRedownload}>{t('common.redownload')}</Button>
+      </div>
+    </div>
+  )
+}
+
+const VideoPlayer = ({
+  video,
+  setLoadSuccess
+}: {
+  video: VideoDownloaded
+  setLoadSuccess: (value: boolean) => void
+}) => {
+  const fetcher = async () => {
+    const file = await FileManager.getFile(video.fileId)
+    if (!file) {
+      throw new Error(`Video file ${video.fileId} not exist.`)
+    }
+    return FileManager.getFilePath(file)
+  }
+  const { data: src, isLoading, error } = useSWRImmutable(`video/file/${video.id}`, fetcher)
+
+  if (error) {
+    setLoadSuccess(false)
+  }
+
+  if (isLoading) {
+    return <Skeleton />
+  }
+
+  return (
+    <video
+      controls
+      className="h-full w-full rounded-2xl bg-black object-contain"
+      onLoadedData={() => setLoadSuccess(true)}
+      onError={() => setLoadSuccess(false)}>
+      <source src={`file://${src}`} type="video/mp4" />
+    </video>
+  )
+}

src/renderer/src/pages/video/settings/ModelSetting.tsx

@@ -0,0 +1,44 @@
+import { Select, SelectItem } from '@heroui/react'
+import { videoModelsMap } from '@renderer/config/models/video'
+import { Model } from '@renderer/types'
+import { useTranslation } from 'react-i18next'
+
+import { SettingItem } from './shared'
+
+export interface ModelSettingProps {
+  providerId: string
+  modelId: string
+  setModelId: (id: string) => void
+}
+
+interface ModelSelectItem extends Model {
+  key: string
+  label: string
+}
+
+export const ModelSetting = ({ providerId, modelId, setModelId }: ModelSettingProps) => {
+  const { t } = useTranslation()
+
+  const items: ModelSelectItem[] = videoModelsMap[providerId]?.map((m: string) => ({ key: m, label: m })) ?? []
+
+  return (
+    <SettingItem>
+      <Select
+        label={t('common.model')}
+        labelPlacement="outside"
+        selectionMode="single"
+        items={items}
+        defaultSelectedKeys={[modelId]}
+        disallowEmptySelection
+        onSelectionChange={(keys) => {
+          if (keys.currentKey) setModelId(keys.currentKey)
+        }}>
+        {(model) => (
+          <SelectItem textValue={model.label}>
+            <span>{model.label}</span>
+          </SelectItem>
+        )}
+      </Select>
+    </SettingItem>
+  )
+}

src/renderer/src/pages/video/settings/OpenAIParamSettings.tsx

@@ -0,0 +1,80 @@
+import { VideoSeconds, VideoSize } from '@cherrystudio/openai/resources'
+import { Select, SelectItem } from '@heroui/react'
+import { OpenAICreateVideoParams } from '@renderer/types/video'
+import { DeepPartial } from 'ai'
+import { useCallback } from 'react'
+import { useTranslation } from 'react-i18next'
+
+import { SettingItem, SettingsGroup } from './shared'
+
+export type OpenAIParamSettingsProps = {
+  params: OpenAICreateVideoParams
+  updateParams: (update: DeepPartial<Omit<OpenAICreateVideoParams, 'type'>>) => void
+}
+
+export const OpenAIParamSettings = ({ params, updateParams }: OpenAIParamSettingsProps) => {
+  const { t } = useTranslation()
+
+  const secondItems = [{ key: '4' }, { key: '8' }, { key: '12' }] as const satisfies { key: VideoSeconds }[]
+  const sizeItems = [
+    { key: '720x1280' },
+    { key: '1280x720' },
+    { key: '1024x1792' },
+    { key: '1792x1024' }
+  ] as const satisfies { key: VideoSize }[]
+
+  const updateSeconds = useCallback(
+    (seconds: VideoSeconds) => {
+      updateParams({ params: { seconds } })
+    },
+    [updateParams]
+  )
+
+  const updateSize = useCallback(
+    (size: VideoSize) => {
+      updateParams({ params: { size } })
+    },
+    [updateParams]
+  )
+
+  return (
+    <SettingsGroup>
+      <SettingItem>
+        <Select
+          label={t('video.seconds')}
+          labelPlacement="outside"
+          selectedKeys={[params.params.seconds ?? '4']}
+          onSelectionChange={(keys) => {
+            if (keys.currentKey) updateSeconds(keys.currentKey as VideoSeconds)
+          }}
+          items={secondItems}
+          selectionMode="single"
+          disallowEmptySelection>
+          {(item) => (
+            <SelectItem key={item.key} textValue={item.key}>
+              <span>{item.key}</span>
+            </SelectItem>
+          )}
+        </Select>
+      </SettingItem>
+      <SettingItem>
+        <Select
+          label={t('video.size')}
+          labelPlacement="outside"
+          selectedKeys={[params.params.size ?? '720x1280']}
+          onSelectionChange={(keys) => {
+            if (keys.currentKey) updateSize(keys.currentKey as VideoSize)
+          }}
+          items={sizeItems}
+          selectionMode="single"
+          disallowEmptySelection>
+          {(item) => (
+            <SelectItem key={item.key} textValue={item.key}>
+              <span>{item.key}</span>
+            </SelectItem>
+          )}
+        </Select>
+      </SettingItem>
+    </SettingsGroup>
+  )
+}

src/renderer/src/pages/video/settings/ProviderSetting.tsx

@@ -0,0 +1,60 @@
+import { Select, SelectItem } from '@heroui/react'
+import { ProviderAvatar } from '@renderer/components/ProviderAvatar'
+import { useProviders } from '@renderer/hooks/useProvider'
+import { Provider, SystemProviderId } from '@renderer/types'
+import { getFancyProviderName } from '@renderer/utils'
+import { Dispatch, SetStateAction } from 'react'
+import { useTranslation } from 'react-i18next'
+
+import { SettingItem } from './shared'
+
+export interface ProviderSettingProps {
+  providerId: string
+  setProviderId: Dispatch<SetStateAction<string>>
+}
+
+interface ProviderSelectItem extends Provider {
+  key: string
+  label: string
+}
+
+export const ProviderSetting = ({ providerId, setProviderId }: ProviderSettingProps) => {
+  const { t } = useTranslation()
+  // Support limited providers.
+  const supportedProviderIds = ['openai'] satisfies SystemProviderId[]
+  const { providers } = useProviders()
+  const items: ProviderSelectItem[] = providers
+    .filter((p) => supportedProviderIds.some((id) => id === p.id))
+    .map((p) => ({ ...p, key: p.id, label: getFancyProviderName(p) }))
+
+  return (
+    <SettingItem>
+      <Select
+        label={t('common.provider')}
+        labelPlacement="outside"
+        selectionMode="single"
+        items={items}
+        defaultSelectedKeys={[providerId]}
+        disallowEmptySelection
+        onSelectionChange={(keys) => {
+          if (keys.currentKey) setProviderId(keys.currentKey)
+        }}
+        renderValue={(items) => {
+          const provider = items[0].data
+          if (!provider) return null
+          return (
+            <div className="flex items-center gap-2">
+              <ProviderAvatar provider={provider} size={16} />
+              <span>{provider.label}</span>
+            </div>
+          )
+        }}>
+        {(provider) => (
+          <SelectItem textValue={provider.label} startContent={<ProviderAvatar provider={provider} size={16} />}>
+            <span>{provider.label}</span>
+          </SelectItem>
+        )}
+      </Select>
+    </SettingItem>
+  )
+}

src/renderer/src/pages/video/settings/shared.tsx

@@ -0,0 +1,15 @@
+import { Divider } from '@heroui/react'
+import { PropsWithChildren } from 'react'
+
+export const SettingsGroup = ({ children }: PropsWithChildren) => {
+  return <div className="mb-4 flex flex-col rounded-2xl border border-foreground-200 p-3">{children}</div>
+}
+
+export const SettingItem = ({ children, divider = false }: PropsWithChildren<{ divider?: boolean }>) => {
+  return (
+    <>
+      <div className="mb-2">{children}</div>
+      {divider && <Divider className="my-2" />}
+    </>
+  )
+}

src/renderer/src/services/ApiService.ts

@@ -10,12 +10,19 @@ import { isDedicatedImageGenerationModel, isEmbeddingModel } from '@renderer/con
 import { getStoreSetting } from '@renderer/hooks/useSettings'
 import i18n from '@renderer/i18n'
 import store from '@renderer/store'
-import type { FetchChatCompletionParams } from '@renderer/types'
+import type { FetchChatCompletionParams, RetrieveVideoContentParams } from '@renderer/types'
 import { Assistant, MCPServer, MCPTool, Model, Provider } from '@renderer/types'
 import type { StreamTextParams } from '@renderer/types/aiCoreTypes'
 import { type Chunk, ChunkType } from '@renderer/types/chunk'
 import { Message } from '@renderer/types/newMessage'
 import { SdkModel } from '@renderer/types/sdk'
+import {
+  CreateVideoParams,
+  CreateVideoResult,
+  RetrieveVideoContentResult,
+  RetrieveVideoParams,
+  RetrieveVideoResult
+} from '@renderer/types/video'
 import { removeSpecialCharactersForTopicName, uuid } from '@renderer/utils'
 import { abortCompletion, readyToAbort } from '@renderer/utils/abortController'
 import { isAbortError } from '@renderer/utils/error'
@@ -397,6 +404,21 @@ export async function fetchGenerate({
   }
 }
 
+export async function createVideo(params: CreateVideoParams): Promise<CreateVideoResult> {
+  const ai = new AiProviderNew(params.provider)
+  return ai.createVideo(params)
+}
+
+export async function retrieveVideo(params: RetrieveVideoParams): Promise<RetrieveVideoResult> {
+  const ai = new AiProviderNew(params.provider)
+  return ai.retrieveVideo(params)
+}
+
+export async function retrieveVideoContent(params: RetrieveVideoContentParams): Promise<RetrieveVideoContentResult> {
+  const ai = new AiProviderNew(params.provider)
+  return ai.retrieveVideoContent(params)
+}
+
 export function hasApiKey(provider: Provider) {
   if (!provider) return false
   if (['ollama', 'lmstudio', 'vertexai', 'cherryai'].includes(provider.id)) return true

src/renderer/src/services/ModelMessageService.ts

@@ -1,6 +1,6 @@
+import { ChatCompletionContentPart, ChatCompletionMessageParam } from '@cherrystudio/openai/resources'
 import { Model } from '@renderer/types'
 import { findLast } from 'lodash'
-import { ChatCompletionContentPart, ChatCompletionMessageParam } from 'openai/resources'
 
 export function processReqMessages(
   model: Model,

src/renderer/src/services/SpanManagerService.ts

@@ -1,4 +1,5 @@
 import { MessageStream } from '@anthropic-ai/sdk/resources/messages/messages'
+import { Stream } from '@cherrystudio/openai/streaming'
 import { loggerService } from '@logger'
 import { SpanEntity, TokenUsage } from '@mcp-trace/trace-core'
 import { cleanContext, endContext, getContext, startContext } from '@mcp-trace/trace-web'
@@ -16,7 +17,6 @@ import { Model, Topic } from '@renderer/types'
 import type { Message } from '@renderer/types/newMessage'
 import { MessageBlockType } from '@renderer/types/newMessage'
 import { SdkRawChunk } from '@renderer/types/sdk'
-import { Stream } from 'openai/streaming'
 
 const logger = loggerService.withContext('SpanManagerService')
 

src/renderer/src/services/__tests__/ApiService.test.ts

@@ -6,6 +6,8 @@ import {
   WebSearchResultBlock,
   WebSearchToolResultError
 } from '@anthropic-ai/sdk/resources/messages'
+import OpenAI from '@cherrystudio/openai'
+import { ChatCompletionChunk } from '@cherrystudio/openai/resources'
 import { FinishReason, MediaModality } from '@google/genai'
 import { FunctionCall } from '@google/genai'
 import AiProvider from '@renderer/aiCore'
@@ -38,8 +40,6 @@ import {
 import { mcpToolCallResponseToGeminiMessage } from '@renderer/utils/mcp-tools'
 import * as McpToolsModule from '@renderer/utils/mcp-tools'
 import { cloneDeep } from 'lodash'
-import OpenAI from 'openai'
-import { ChatCompletionChunk } from 'openai/resources'
 import { beforeEach, describe, expect, it, vi } from 'vitest'
 // Mock the ApiClientFactory
 vi.mock('@renderer/aiCore/legacy/clients/ApiClientFactory', () => ({

src/renderer/src/services/__tests__/ModelMessageService.test.ts

@@ -1,5 +1,5 @@
+import { ChatCompletionMessageParam } from '@cherrystudio/openai/resources'
 import type { Model } from '@renderer/types'
-import { ChatCompletionMessageParam } from 'openai/resources'
 import { describe, expect, it } from 'vitest'
 
 import { processReqMessages } from '../ModelMessageService'

src/renderer/src/store/index.ts

@@ -30,6 +30,7 @@ import settings from './settings'
 import shortcuts from './shortcuts'
 import tabs from './tabs'
 import translate from './translate'
+import video from './video'
 import websearch from './websearch'
 
 const logger = loggerService.withContext('Store')
@@ -58,14 +59,15 @@ const rootReducer = combineReducers({
   inputTools: inputToolsReducer,
   translate,
   ocr,
-  note
+  note,
+  video
 })
 
 const persistedReducer = persistReducer(
   {
     key: 'cherry-studio',
     storage,
-    version: 162,
+    version: 163,
     blacklist: ['runtime', 'messages', 'messageBlocks', 'tabs'],
     migrate
   },

src/renderer/src/store/messageBlock.ts

@@ -1,10 +1,10 @@
 import { WebSearchResultBlock } from '@anthropic-ai/sdk/resources'
+import type OpenAI from '@cherrystudio/openai'
 import type { GroundingMetadata } from '@google/genai'
 import { createEntityAdapter, createSelector, createSlice, type PayloadAction } from '@reduxjs/toolkit'
 import { AISDKWebSearchResult, Citation, WebSearchProviderResponse, WebSearchSource } from '@renderer/types'
 import type { CitationMessageBlock, MessageBlock } from '@renderer/types/newMessage'
 import { MessageBlockType } from '@renderer/types/newMessage'
-import type OpenAI from 'openai'
 
 import type { RootState } from './index' // 确认 RootState 从 store/index.ts 导出
 

src/renderer/src/store/migrate.ts

@@ -2667,6 +2667,20 @@ const migrateConfig = {
       logger.error('migrate 162 error', error as Error)
       return state
     }
+  },
+  '163': (state: RootState) => {
+    try {
+      if (state.settings && state.settings.sidebarIcons) {
+        if (!state.settings.sidebarIcons.visible.includes('video')) {
+          state.settings.sidebarIcons.visible = [...state.settings.sidebarIcons.visible, 'video']
+        }
+      }
+      state.video.videoMap = {}
+      return state
+    } catch (error) {
+      logger.error('migrate 161 error', error as Error)
+      return state
+    }
   }
 }
 

src/renderer/src/store/video.ts

@@ -0,0 +1,78 @@
+import { loggerService } from '@logger'
+import { createSlice, PayloadAction } from '@reduxjs/toolkit'
+import { Video } from '@renderer/types/video'
+
+const logger = loggerService.withContext('Store:video')
+
+export interface VideoState {
+  /** Provider ID to videos */
+  videoMap: Record<string, Video[] | undefined>
+}
+
+const initialState: VideoState = {
+  videoMap: {}
+}
+
+const videoSlice = createSlice({
+  name: 'video',
+  initialState,
+  reducers: {
+    addVideo: (state: VideoState, action: PayloadAction<{ providerId: string; video: Video }>) => {
+      const { providerId, video } = action.payload
+      if (state.videoMap[providerId]) {
+        state.videoMap[providerId].unshift(video)
+      } else {
+        state.videoMap[providerId] = [video]
+      }
+    },
+    removeVideo: (state: VideoState, action: PayloadAction<{ providerId: string; videoId: string }>) => {
+      const { providerId, videoId } = action.payload
+      const videos = state.videoMap[providerId]
+      state.videoMap[providerId] = videos?.filter((c) => c.id !== videoId)
+    },
+    updateVideo: (
+      state: VideoState,
+      action: PayloadAction<{ providerId: string; update: Partial<Omit<Video, 'status'>> & { id: string } }>
+    ) => {
+      const { providerId, update } = action.payload
+      const videos = state.videoMap[providerId]
+      if (videos) {
+        let video = videos.find((v) => v.id === update.id)
+        if (video) {
+          video = { ...video, ...update }
+        } else {
+          logger.error(`Video with id ${update.id} not found in ${providerId}`)
+        }
+      } else {
+        logger.error(`Videos with Provider ${providerId} is undefined.`)
+      }
+    },
+    setVideo: (state: VideoState, action: PayloadAction<{ providerId: string; video: Video }>) => {
+      const { providerId, video } = action.payload
+      if (state.videoMap[providerId]) {
+        const index = state.videoMap[providerId].findIndex((v) => v.id === video.id)
+        if (index !== -1) {
+          state.videoMap[providerId][index] = video
+        } else {
+          state.videoMap[providerId].push(video)
+        }
+      } else {
+        state.videoMap[providerId] = [video]
+      }
+    },
+    setVideos: (state: VideoState, action: PayloadAction<{ providerId: string; videos: Video[] }>) => {
+      const { providerId, videos } = action.payload
+      state.videoMap[providerId] = videos
+    }
+  }
+})
+
+export const {
+  addVideo: addVideoAction,
+  removeVideo: removeVideoAction,
+  updateVideo: updateVideoAction,
+  setVideo: setVideoAction,
+  setVideos: setVideosAction
+} = videoSlice.actions
+
+export default videoSlice.reducer

src/renderer/src/trace/dataHandler/StreamHandler.ts

@@ -1,8 +1,8 @@
+import { OpenAI } from '@cherrystudio/openai'
+import { Stream } from '@cherrystudio/openai/streaming'
 import { TokenUsage } from '@mcp-trace/trace-core'
 import { Span } from '@opentelemetry/api'
 import { endSpan } from '@renderer/services/SpanManagerService'
-import { OpenAI } from 'openai'
-import { Stream } from 'openai/streaming'
 
 export class StreamHandler {
   private topicId: string

src/renderer/src/types/file.ts

@@ -1,6 +1,6 @@
+import type OpenAI from '@cherrystudio/openai'
 import type { File } from '@google/genai'
 import type { FileSchema } from '@mistralai/mistralai/models/components'
-import type OpenAI from 'openai'
 
 export type RemoteFile =
   | {
@@ -127,6 +127,10 @@ export type ImageFileMetadata = FileMetadata & {
   type: FileTypes.IMAGE
 }
 
+export type VideoFileMetadata = FileMetadata & {
+  type: FileTypes.VIDEO
+}
+
 export type PdfFileMetadata = FileMetadata & {
   ext: '.pdf'
 }

src/renderer/src/types/index.ts

@@ -1,7 +1,7 @@
 import type { LanguageModelV2Source } from '@ai-sdk/provider'
 import type { WebSearchResultBlock } from '@anthropic-ai/sdk/resources'
+import type OpenAI from '@cherrystudio/openai'
 import type { GenerateImagesConfig, GroundingMetadata, PersonGeneration } from '@google/genai'
-import type OpenAI from 'openai'
 import type { CSSProperties } from 'react'
 
 export * from './file'
@@ -22,6 +22,7 @@ export * from './mcp'
 export * from './notification'
 export * from './ocr'
 export * from './provider'
+export * from './video'
 
 export type Assistant = {
   id: string
@@ -535,6 +536,7 @@ export type SidebarIcon =
   | 'files'
   | 'code_tools'
   | 'notes'
+  | 'video'
 
 export type ExternalToolResult = {
   mcpTools?: MCPTool[]

src/renderer/src/types/newMessage.ts

@@ -1,5 +1,5 @@
+import type { CompletionUsage } from '@cherrystudio/openai/resources'
 import type { ProviderMetadata } from 'ai'
-import type { CompletionUsage } from 'openai/resources'
 
 import type {
   Assistant,

src/renderer/src/types/sdk.ts

@@ -11,6 +11,9 @@ import { MessageStream } from '@anthropic-ai/sdk/resources/messages/messages'
 import AnthropicVertex from '@anthropic-ai/vertex-sdk'
 import type { BedrockClient } from '@aws-sdk/client-bedrock'
 import type { BedrockRuntimeClient } from '@aws-sdk/client-bedrock-runtime'
+import OpenAI, { AzureOpenAI } from '@cherrystudio/openai'
+import { ChatCompletionContentPartImage } from '@cherrystudio/openai/resources'
+import { Stream } from '@cherrystudio/openai/streaming'
 import {
   Content,
   CreateChatParameters,
@@ -21,9 +24,6 @@ import {
   SendMessageParameters,
   Tool
 } from '@google/genai'
-import OpenAI, { AzureOpenAI } from 'openai'
-import { ChatCompletionContentPartImage } from 'openai/resources'
-import { Stream } from 'openai/streaming'
 
 import { EndpointType } from './index'
 

src/renderer/src/types/video.ts

@@ -0,0 +1,162 @@
+import OpenAI from '@cherrystudio/openai'
+
+import { Provider } from './provider'
+
+// Only OpenAI (Responses) is supported for now.
+export type VideoEndpointType = 'openai'
+export type VideoStatus = 'queued' | 'in_progress' | 'completed' | 'downloading' | 'downloaded' | 'failed'
+
+interface VideoBase {
+  readonly id: string
+  readonly type: VideoEndpointType
+  name: string
+  readonly providerId: string
+  prompt: string
+  /**
+   * Represents the possible states of a video generation or download process.
+   *
+   * - `queued`: The video task has been submitted and is waiting to be processed.
+   * - `in_progress`: The video is currently being generated.
+   * - `completed`: The video has been successfully generated and is ready for download.
+   * - `downloading`: The video content is being downloaded to local storage.
+   * - `downloaded`: The video has been fully downloaded and is available locally.
+   * - `failed`: The video task encountered an error and could not be completed.
+   */
+  readonly status: VideoStatus
+}
+
+interface OpenAIVideoBase {
+  readonly type: 'openai'
+  metadata: OpenAI.Videos.Video
+}
+
+export interface VideoQueued extends VideoBase {
+  readonly status: 'queued'
+}
+
+export interface VideoInProgress extends VideoBase {
+  readonly status: 'in_progress'
+  /** integer percent */
+  progress: number
+}
+export interface VideoCompleted extends VideoBase {
+  readonly status: 'completed'
+  /** When generation completed, firstly try to retrieve thumbnail. */
+  thumbnail: string | null
+}
+
+export interface VideoDownloading extends VideoBase {
+  readonly status: 'downloading'
+  thumbnail: string | null
+  /** integer percent */
+  progress: number
+}
+export interface VideoDownloaded extends VideoBase {
+  readonly status: 'downloaded'
+  thumbnail: string | null
+  /** Managed by fileManager */
+  fileId: string
+}
+
+export interface VideoFailedBase extends VideoBase {
+  readonly status: 'failed'
+  error: unknown
+}
+
+export interface OpenAIVideoQueued extends VideoQueued, OpenAIVideoBase {}
+export interface OpenAIVideoInProgress extends VideoInProgress, OpenAIVideoBase {}
+export interface OpenAIVideoCompleted extends VideoCompleted, OpenAIVideoBase {}
+export interface OpenAIVideoDownloading extends VideoDownloading, OpenAIVideoBase {}
+export interface OpenAIVideoDownloaded extends VideoDownloaded, OpenAIVideoBase {}
+export interface OpenAIVideoFailed extends VideoFailedBase, OpenAIVideoBase {
+  error: OpenAI.Videos.Video['error']
+}
+
+export type VideoFailed = OpenAIVideoFailed
+
+export type OpenAIVideo =
+  | OpenAIVideoQueued
+  | OpenAIVideoInProgress
+  | OpenAIVideoCompleted
+  | OpenAIVideoDownloading
+  | OpenAIVideoDownloaded
+  | OpenAIVideoFailed
+
+export type Video = OpenAIVideo
+
+// Create Video
+interface CreateVideoBaseParams {
+  type: VideoEndpointType
+  provider: Provider
+}
+
+export interface OpenAICreateVideoParams extends CreateVideoBaseParams {
+  type: 'openai'
+  params: OpenAI.VideoCreateParams
+  options?: OpenAI.RequestOptions
+}
+
+export type CreateVideoParams = OpenAICreateVideoParams
+
+interface CreateVideoBaseResult {
+  type: VideoEndpointType
+  video: unknown
+}
+
+export interface OpenAICreateVideoResult extends CreateVideoBaseResult {
+  type: 'openai'
+  video: OpenAI.Videos.Video
+}
+
+export type CreateVideoResult = OpenAICreateVideoResult
+
+// Retrieve Video
+interface RetrieveVideoBaseParams {
+  type: VideoEndpointType
+  provider: Provider
+}
+
+export interface OpenAIRetrieveVideoParams extends RetrieveVideoBaseParams {
+  type: 'openai'
+  videoId: string
+  options?: OpenAI.RequestOptions
+}
+
+export type RetrieveVideoParams = OpenAIRetrieveVideoParams
+
+interface RetrieveVideoBaseResult {
+  type: VideoEndpointType
+}
+
+export interface OpenAIRetrieveVideoResult extends RetrieveVideoBaseResult {
+  type: 'openai'
+  video: OpenAI.Videos.Video
+}
+
+export type RetrieveVideoResult = OpenAIRetrieveVideoResult
+
+// Retrieve Video Content
+interface RetrieveVideoContentBaseParams {
+  type: VideoEndpointType
+  provider: Provider
+}
+
+export interface OpenAIRetrieveVideoContentParams extends RetrieveVideoContentBaseParams {
+  type: 'openai'
+  videoId: string
+  query?: OpenAI.Videos.VideoDownloadContentParams
+  options?: OpenAI.RequestOptions
+}
+
+export type RetrieveVideoContentParams = OpenAIRetrieveVideoContentParams
+
+interface RetrieveVideoContentBaseResult {
+  type: VideoEndpointType
+}
+
+export interface OpenAIRetrieveVideoContentResult extends RetrieveVideoContentBaseResult {
+  type: 'openai'
+  response: Response
+}
+
+export type RetrieveVideoContentResult = OpenAIRetrieveVideoContentResult

src/renderer/src/utils/deepUpdate.ts

@@ -0,0 +1,35 @@
+import { DeepPartial } from 'ai'
+import { cloneDeep } from 'lodash'
+
+/**
+ * Deeply updates an object, allowing undefined to overwrite existing properties, without using `any`
+ * @param target Original object
+ * @param update Update object (may contain undefined)
+ * @returns New object
+ */
+export function deepUpdate<T extends object>(target: T, update: DeepPartial<T>): T {
+  const result = cloneDeep(target)
+  for (const key in update) {
+    if (Object.hasOwn(update, key)) {
+      // @ts-ignore it's runtime safe
+      const prev = result[key]
+      const next = update[key]
+
+      if (
+        next &&
+        typeof next === 'object' &&
+        !Array.isArray(next) &&
+        prev &&
+        typeof prev === 'object' &&
+        !Array.isArray(prev)
+      ) {
+        // @ts-ignore it's runtime safe
+        result[key] = deepUpdate(prev, next as any)
+      } else {
+        // @ts-ignore it's runtime safe
+        result[key] = next
+      }
+    }
+  }
+  return result
+}

src/renderer/src/utils/mcp-tools.ts

@@ -1,4 +1,11 @@
 import { ContentBlockParam, MessageParam, ToolUnion, ToolUseBlock } from '@anthropic-ai/sdk/resources'
+import OpenAI from '@cherrystudio/openai'
+import {
+  ChatCompletionContentPart,
+  ChatCompletionMessageParam,
+  ChatCompletionMessageToolCall,
+  ChatCompletionTool
+} from '@cherrystudio/openai/resources'
 import { Content, FunctionCall, Part, Tool, Type as GeminiSchemaType } from '@google/genai'
 import { loggerService } from '@logger'
 import { isFunctionCallingModel, isVisionModel } from '@renderer/config/models'
@@ -21,13 +28,6 @@ import { ChunkType } from '@renderer/types/chunk'
 import { AwsBedrockSdkMessageParam, AwsBedrockSdkTool, AwsBedrockSdkToolCall } from '@renderer/types/sdk'
 import { t } from 'i18next'
 import { nanoid } from 'nanoid'
-import OpenAI from 'openai'
-import {
-  ChatCompletionContentPart,
-  ChatCompletionMessageParam,
-  ChatCompletionMessageToolCall,
-  ChatCompletionTool
-} from 'openai/resources'
 
 import { isToolUseModeFunction } from './assistant'
 import { convertBase64ImageToAwsBedrockFormat } from './aws-bedrock-utils'

src/renderer/src/utils/model/video.ts

@@ -0,0 +1,7 @@
+import { VideoModel } from '@cherrystudio/openai/resources'
+import { videoModelsMap } from '@renderer/config/models/video'
+
+// Only for openai, use hard-encoded values
+export const isVideoModel = (modelId: string): modelId is VideoModel => {
+  return videoModelsMap.openai.some((v) => v === modelId)
+}

yarn.lock

@@ -2665,6 +2665,23 @@ __metadata:
   languageName: unknown
   linkType: soft
 
+"@cherrystudio/openai@npm:6.3.0-fork.1, openai@npm:@cherrystudio/openai@6.3.0-fork.1":
+  version: 6.3.0-fork.1
+  resolution: "@cherrystudio/openai@npm:6.3.0-fork.1"
+  peerDependencies:
+    ws: ^8.18.0
+    zod: ^3.25 || ^4.0
+  peerDependenciesMeta:
+    ws:
+      optional: true
+    zod:
+      optional: true
+  bin:
+    openai: bin/cli
+  checksum: 10c0/dc8c5555aa6d12cd47586efc70175ec1bf7112c1f737b28d86d26e9666535d9e87d1c4811a069d11c324bab56f4dae242d7266efa88359d0926c7059240d24cc
+  languageName: node
+  linkType: hard
+
 "@chevrotain/cst-dts-gen@npm:11.0.3":
   version: 11.0.3
   resolution: "@chevrotain/cst-dts-gen@npm:11.0.3"
@@ -14286,6 +14303,7 @@ __metadata:
     "@cherrystudio/embedjs-ollama": "npm:^0.1.31"
     "@cherrystudio/embedjs-openai": "npm:^0.1.31"
     "@cherrystudio/extension-table-plus": "workspace:^"
+    "@cherrystudio/openai": "npm:6.3.0-fork.1"
     "@dnd-kit/core": "npm:^6.3.1"
     "@dnd-kit/modifiers": "npm:^9.0.0"
     "@dnd-kit/sortable": "npm:^10.0.0"
@@ -14466,7 +14484,6 @@ __metadata:
     notion-helper: "npm:^1.3.22"
     npx-scope-finder: "npm:^1.2.0"
     officeparser: "npm:^4.2.0"
-    openai: "patch:openai@npm%3A5.12.2#~/.yarn/patches/openai-npm-5.12.2-30b075401c.patch"
     os-proxy-config: "npm:^1.1.2"
     oxlint: "npm:^1.22.0"
     oxlint-tsgolint: "npm:^0.2.0"
@@ -24491,23 +24508,6 @@ __metadata:
   languageName: node
   linkType: hard
 
-"openai@patch:openai@npm%3A5.12.2#~/.yarn/patches/openai-npm-5.12.2-30b075401c.patch":
-  version: 5.12.2
-  resolution: "openai@patch:openai@npm%3A5.12.2#~/.yarn/patches/openai-npm-5.12.2-30b075401c.patch::version=5.12.2&hash=ad5d10"
-  peerDependencies:
-    ws: ^8.18.0
-    zod: ^3.23.8
-  peerDependenciesMeta:
-    ws:
-      optional: true
-    zod:
-      optional: true
-  bin:
-    openai: bin/cli
-  checksum: 10c0/2964a1c88a98cf169c9b73e8cd6776c03c8f3103fee30961c6953e5d995ad57a697e2179615999356809349186df6496abae105928ff7ce0229e5016dec87cb3
-  languageName: node
-  linkType: hard
-
 "openapi-types@npm:^12.1.3":
   version: 12.1.3
   resolution: "openapi-types@npm:12.1.3"