* refactor(ApiKeyList): add a popup for api key list
- ApiKeyList for key management
- ApiKeyListPopup triggerred by a button
- Move formatApiKeys to utils for better reuse
- Simplify apikey related states in ProviderSettings for better
integration with ApiKeyList
- Modify `updateProvider` to accept partial updates
- Update api key placeholder
* fix: strict type
* refactor: support websearch provider
* refactor: remove ApiCheckPopup
* refactor: simplify interfaces for ProviderSetting and WebSearchProviderSetting
* fix: sync input api key between sub-pages, futher simplification
* fix: bold title
* refactor: extract status icon colors
* refactor: add a status indicator to input box on error, update type definitions
* refactor: further simplification, make data flow clearer
* feat: support api key list for preprocess settings
* refactor: better naming, less confusion
* test: add unit tests for asyncInitializer and copy utilities
- Add tests for asyncInitializer class functionality
- Add tests for clipboard copy operations
* refactor(test): improve copy.test.ts structure and maintainability
- Remove complex shared testCopyFunction in favor of individual test cases
- Simplify mock cleanup by removing redundant afterEach
- Split test scenarios into focused, independent test cases
- Improve test readability with clear Chinese comments
- Maintain full test coverage while following TEST_UTILS.md guidelines
- Fix minor formatting in asyncInitializer.test.ts
* test: remove unnecessary test cases
- Remove AsyncInitializer type support test
- Remove maintain separate instances test
- These tests verify language features rather than business logic
* refactor(test): reorganize copy and export test structure
Restructure test organization based on PR review feedback:
- Move export functionality tests from copy.test.ts to export.test.ts
- Remove unnecessary "clipboard API not available" test
- Merge duplicate empty content tests for better coverage
- Add boundary tests for special characters and Markdown formatting
- Fix ESLint formatting issues
Test responsibilities are now clearer:
- copy.test.ts: Focus on clipboard operations (8 tests)
- export.test.ts: Focus on content conversion and edge cases
* fix(test): correct markdown formatting test for list items
Fix the regex pattern to properly handle markdown list items.
Replace with separate patterns to avoid removing
the dash from list items incorrectly.
* fix(test): format prettier style for markdown test
* test: add comprehensive tests for CopyIcon and MinAppIcon components
- Add tests for CopyIcon covering default rendering, className merging, and prop passing
- Add tests for MinAppIcon covering default props, custom size, sidebar mode, styles, and edge cases
- Include snapshot tests for both components
* fix: update test snapshots after component styling changes
Update snapshots for CopyIcon and MinAppIcon components to match current
styled-components implementation (replaces inline styles with generated classes).
* refactor: simplify icon component tests based on PR review feedback
- CopyIcon: replace multiple redundant tests with single snapshot test
- MinAppIcon: remove duplicate test that overlaps with snapshot test
- Keep essential business logic tests for MinAppIcon (sidebar behavior, null return)
- Update test snapshots accordingly
- Changed the default width from 1080 to 960 and height from 670 to 600 for the main window.
- Adjusted minimum width and height settings to match the new defaults, enhancing compatibility with various screen sizes.
* feat(settings): add option to disable hardware acceleration
- Introduced a new setting to allow users to disable hardware acceleration.
- Added corresponding IPC channel and configuration management methods.
- Updated UI components to reflect the new setting and prompt for app restart.
- Localized confirmation messages for hardware acceleration changes in multiple languages.
* fix(settings): add delay before relaunching app after disabling hardware acceleration
- Introduced a 500ms delay before the application relaunches to ensure settings are applied correctly.
- This change improves user experience by allowing time for the setting to take effect before the app restarts.
* fix lint
* fix(settings): handle errors when disabling hardware acceleration
- Wrapped the hardware acceleration disabling function in a try-catch block to manage potential errors.
- Added user feedback through an error message if the operation fails, improving overall robustness.
* feat: enhance citation handling for Perplexity web search results
- Implemented formatting for Perplexity citations in MainTextBlock, including data-citation attributes.
- Updated citation processing in message store and thunk to support new citation structure.
- Added utility functions for link completion based on web search results.
- Enhanced tests to verify correct handling of Perplexity citations and links.
* refactor: streamline chunk processing in OpenAIApiClient
- Replaced single choice handling with a loop to process all choices in the chunk.
- Improved handling of content sources, ensuring fallback mechanisms are in place for delta and message fields.
- Enhanced tool call processing to accommodate missing function names and arguments.
- Maintained existing functionality for web search data and reasoning content processing.
* fix: improve citation handling and web search integration
- Enhanced citation formatting to support legacy data compatibility in messageBlock.ts.
- Updated messageThunk.ts to manage main text block references and citation updates more effectively.
- Removed unnecessary web search flag and streamlined block processing logic.
* fix: improve citation transforms to skip code blocks
- Add withCitationTags for better code structure
- Add tests
- Remove outdated code
- The Citation type in @renderer/types/index.ts is not referenced anywhere, so removed
- Move the actual Citation type from @renderer/pages/home/Messages/CitationsList.tsx to @renderer/types/index.ts
- Allow text selecting in tooltip
* test: update tests
* refactor(messageThunk): streamline citation handling in response processing
- Removed redundant citation block source retrieval during text chunk processing.
- Updated citation references handling to ensure proper inclusion only when available.
- Simplified the logic for managing citation references in both streaming and final text updates.
* refactor: simplify determineCitationSource for backward compatibility
---------
Co-authored-by: one <wangan.cs@gmail.com>
* refactor(messageThunk): streamline loading state management for topics
- Reintroduced the handleChangeLoadingOfTopic function to manage loading states more effectively.
- Updated thunk implementations to ensure loading state is correctly set after message processing.
- Removed commented-out code for clarity and maintainability.
* fix(messageThunk): ensure loading state is managed correctly after message sending
- Added a finally block to guarantee that the loading state is updated after the sendMessage thunk execution.
- Removed commented-out code for improved clarity and maintainability.
* perf(CodePreview: virtual list for shiki code block
- move code highlighting to a hook
- use @tanstack/react-virtual dynamic list for CodePreview
- highlight visible items on demand
* refactor: change absolute position to relative position
* refactor: update shiki styles, set scrollbar color for shiki themes
* feat(provider): add NewAPI provider
* feat(providers): Enhance New API model discovery and configuration
This commit refactors the model fetching mechanism for the "New API" provider to improve user experience and support more detailed model information.
The `NewAPIClient` now fetches models directly from the `/models` endpoint, which provides richer metadata, including a new `supported_endpoint_types` field.
Key changes:
- The "Edit Models" popup now automatically adds a model if its `supported_endpoint_types` are provided by the API, using the first available type.
- The manual "Add Model" popup is now a fallback for models that do not declare their endpoint types.
- A new `NewApiModel` type is introduced to handle the structured API response.
- Added support for the `jina-rerank` endpoint type.
* chore(store): update version to 119 and adjust migration function for state management
* fix: adjust label column flex for New API provider in ModelEditContent and NewApiAddModelPopup
* feat: Implement batch adding for New API provider
* feat: Add useDynamicLabelWidth hook for adaptive label widths in forms and fix localization typos
* fix: update dependencies in various components to include translation function
---------
Co-authored-by: 自由的世界人 <3196812536@qq.com>
- Updated the English and Chinese README files to improve layout and add new language options.
- Removed the Japanese README file as part of the documentation cleanup.
- Enhanced badge visibility and adjusted image sizes for better presentation.
- Added GitHub statistics section to provide insights into project activity.
- Add tests for IndicatorLight component covering size, color conversion, shadow, and animation props
- Add tests for Spinner component with proper motion/react mocking
- Include snapshot tests for both components
- Added margin to the modal close button for improved spacing.
- Removed unnecessary divider in OCR settings for a cleaner layout.
- Set a minimum width for the search max result title to enhance alignment and readability.
- Moved the PH8 provider configuration within the INITIAL_PROVIDERS array for better structure.
- Updated migration logic to ensure the PH8 provider is added and positioned correctly in the state during configuration migration.
- Removed redundant code related to provider initialization in the migration process, streamlining the overall logic.
- Added new localization strings for provider API key management, including confirmation and error messages for existing keys.
- Updated English, Japanese, Russian, Simplified Chinese, and Traditional Chinese localization files to reflect these changes, enhancing user experience and clarity in API key operations.
- Removed the left menu and integrated settings directly into the main panel for a more cohesive user experience.
- Introduced a toggle for advanced settings, allowing users to expand or collapse additional configuration options.
- Updated layout and styling for improved usability, including adjustments to padding and margins.
- Enhanced scroll behavior for the advanced settings section to ensure visibility when expanded.
- Minor adjustments to component imports and state management for better performance and clarity.
- Introduced PH8 provider with configuration and logo.
- Updated SYSTEM_MODELS to include PH8 models.
- Added PH8 to internationalization files for multiple languages.
- Implemented migration logic to integrate PH8 into the existing provider structure.
Co-authored-by: jack.li <jack.li@enflame-tech.com>
* fix(dexieError): initialize database connection before fetching phrases
- Added an `init` method to the `QuickPhraseService` to ensure the Dexie database is opened before retrieving all quick phrases.
- Updated the `getAll` method to call the `init` method, improving reliability in data retrieval.
* fix(QuickPhraseService): ensure database initialization before updating phrases
- Added calls to the `init` method in the `update` and `updateOrder` methods to guarantee the database connection is established before performing updates, enhancing data integrity and reliability.
* fix(QuickPhraseService): prevent multiple database initializations
- Added a static flag to ensure the database initialization occurs only once, preventing redundant calls to the `init` method and improving performance.
* feat(SelectionAssistant): add macOS support and process trust handling
- Updated the selection assistant to support macOS, including new IPC channels for process trust verification.
- Enhanced the SelectionService to check for accessibility permissions on macOS before starting the service.
- Added user interface elements to guide macOS users in granting necessary permissions.
- Updated localization files to reflect macOS support and provide relevant user instructions.
- Refactored selection-related configurations to accommodate both Windows and macOS environments.
* feat(SelectionService): update toolbar window settings for macOS and Windows
- Set the toolbar window to be hidden in Mission Control and accept the first mouse click on macOS.
- Adjusted visibility settings for the toolbar window to ensure it appears correctly on all workspaces, including full-screen mode.
- Refactored the MacProcessTrustHintModal component to improve layout and styling of buttons in the modal footer.
* feat(SelectionToolbar): enhance styling and layout of selection toolbar components
* feat(SelectionService): enhance toolbar window settings and refactor position calculation
* feat(SelectionToolbar): update button padding and add last button padding for improved layout
* chore(dependencies): update selection-hook to version 1.0.2 and refine build file exclusions in electron-builder.yml
* feat(SelectionService): center action window on screen when not following toolbar
* fix(SelectionService): implement workaround to prevent other windows from bringing the app to front on macOS when action window is closed
* fix(SelectionService): refine macOS workaround to prevent other windows from bringing the app to front when action window is closed; update selection-toolbar logo padding in styles
* fix(SelectionService): implement macOS toolbar reload to clear hover status; optimize display retrieval logic
* fix(SelectionService): update macOS toolbar hover status handling by sending mouseMove event instead of reloading the window
* chore: update selection-hook dependency to version 1.0.3 in package.json and yarn.lock
* fix(SelectionService): improve toolbar visibility handling on macOS and ensure focusability of other windows when hiding the toolbar
* chore: update selection-hook dependency to version 1.0.4 in package.json and yarn.lock
---------
Co-authored-by: Teo <cheesen.xu@gmail.com>
* fix(messageThunk): handle missing user message in response creation
* fix(i18n): add missing user message translations
* fix(messageThunk): show error popup for missing user message instead of creating error block
* fix(messageThunk): validate askId and show error popup for missing user message
---------
Co-authored-by: suyao <sy20010504@gmail.com>
fix(migrate): enable stream output for existing assistants in migration process
- Updated the migration logic to set the default streamOutput setting to true for assistants that do not have this property defined, enhancing the user experience by ensuring consistent behavior across all assistants.
* feat(SelectionAssistant): add macOS support and process trust handling
- Updated the selection assistant to support macOS, including new IPC channels for process trust verification.
- Enhanced the SelectionService to check for accessibility permissions on macOS before starting the service.
- Added user interface elements to guide macOS users in granting necessary permissions.
- Updated localization files to reflect macOS support and provide relevant user instructions.
- Refactored selection-related configurations to accommodate both Windows and macOS environments.
* feat(SelectionService): update toolbar window settings for macOS and Windows
- Set the toolbar window to be hidden in Mission Control and accept the first mouse click on macOS.
- Adjusted visibility settings for the toolbar window to ensure it appears correctly on all workspaces, including full-screen mode.
- Refactored the MacProcessTrustHintModal component to improve layout and styling of buttons in the modal footer.
* feat(SelectionToolbar): enhance styling and layout of selection toolbar components
* feat(SelectionService): enhance toolbar window settings and refactor position calculation
* feat(SelectionToolbar): update button padding and add last button padding for improved layout
* chore(dependencies): update selection-hook to version 1.0.2 and refine build file exclusions in electron-builder.yml
* feat(SelectionService): center action window on screen when not following toolbar
* fix(SelectionService): implement workaround to prevent other windows from bringing the app to front on macOS when action window is closed
* fix(SelectionService): refine macOS workaround to prevent other windows from bringing the app to front when action window is closed; update selection-toolbar logo padding in styles
* fix(SelectionService): implement macOS toolbar reload to clear hover status; optimize display retrieval logic
* fix(SelectionService): update macOS toolbar hover status handling by sending mouseMove event instead of reloading the window
* chore: update selection-hook dependency to version 1.0.3 in package.json and yarn.lock
* fix(SelectionService): improve toolbar visibility handling on macOS and ensure focusability of other windows when hiding the toolbar
---------
Co-authored-by: Teo <cheesen.xu@gmail.com>
refactor(AboutSettings): streamline test channel change handling
- Moved the test channel change logic into a dedicated function to improve clarity and maintainability.
- Removed the useEffect hook that was previously monitoring changes, simplifying the component's structure.
* feat: 添加代码检查插件支持到 Electron 配置中，并更新依赖项
* test: Update snapshots to reflect the latest changes in component rendering
- Updated snapshots for DragableList, Scrollbar, CitationTooltip, Markdown, Table, and ThinkingBlock components by adding new data attributes to support debugging and testing.
- Ensured snapshots are consistent with the latest component rendering, improving test accuracy and reliability.
* test: 更新快照以反映组件渲染的最新变化
- 更新了 DragableList、Scrollbar、CitationTooltip、Markdown、Table 和 ThinkingBlock 组件的快照，移除了多余的数据属性以简化调试和测试。
- 确保快照与最新的组件渲染一致，提高了测试的准确性和可靠性。
* refactor(ProviderSettings): streamline API key management and enhance user experience
- Refactored the handleProvidersProtocolUrl function to simplify API key handling and improve navigation logic.
- Updated the useProviders hook to maintain consistency in provider management.
- Enhanced the ApiKeyList component with improved state handling and user feedback for API key validation.
- Updated localization files to reflect changes in API key management and user interactions.
- Improved styling and layout for better visual consistency across provider settings.
* fix(ProviderSettings): enhance confirmation modal title with provider name
- Updated the confirmation modal title in the ProvidersList component to include the provider's display name, improving clarity for users during API key management.
* update info
* udpate line
* update line
* feat(Protocol): add custom protocol handling for Cherry Studio
- Introduced a new protocol handler for 'cherrystudio' in the Electron app, allowing the application to respond to custom URL schemes.
- Updated the electron-builder configuration to register the 'cherrystudio' protocol.
- Enhanced the main application logic to handle incoming protocol URLs effectively, improving user experience when launching the app via custom links.
* feat(ProviderSettings): enhance provider data handling with optional fields
- Updated the handleProviderAddKey function to accept optional 'name' and 'type' fields for providers, improving flexibility in provider management.
- Adjusted the API key handling logic to utilize these new fields, ensuring a more comprehensive provider configuration.
- Enhanced the URL schema documentation to reflect the changes in provider data structure.
* delete apikeylist
* restore apiService
* support utf8
* feat(Protocol): improve URL handling for macOS and Windows
- Added caching for the URL received when the app is already running on macOS, ensuring it is processed correctly.
- Updated the URL processing logic in handleProvidersProtocolUrl to replace characters for proper decoding.
- Simplified base64 decoding in ProviderSettings to enhance readability and maintainability.
* fix start in macOS
* format code
* fix(ProviderSettings): validate provider data before adding
- Added validation to ensure 'id', 'newApiKey', and 'baseUrl' are present before proceeding with provider addition.
- Implemented error handling to notify users of invalid data and redirect them to the provider settings page.
* feat(Protocol): enhance URL processing for versioning
- Updated the URL handling logic in handleProvidersProtocolUrl to support versioning by extracting the 'v' parameter.
- Added logging for version 1 to facilitate future enhancements in handling different protocol versions.
- Improved the processing of the 'data' parameter for better compatibility with the updated URL schema.
* feat(i18n): add provider API key management translations for Japanese, Russian, and Traditional Chinese
- Introduced new translations for API key management features, including confirmation prompts and error messages related to provider API keys.
- Enhanced user experience by providing localized strings for adding, updating, and validating API keys across multiple languages.
---------
Co-authored-by: rcadmin <rcadmin@rcadmins-MacBook-Pro-4.local>
* feat(AppUpdater): integrate User-Agent generation for autoUpdater requests; add systemInfo utility module
* feat(systemInfo): enhance macOS version handling using macos-release package for improved accuracy; update package.json and yarn.lock to include macos-release and opendal dependencies
* feat: integrate PPIO OAuth login support
Add OAuth authentication support for PPIO provider with complete integration:
- Add PPIO OAuth configuration and client ID
- Implement oauthWithPPIO authentication flow
- Add PPIO to OAuth and charge-supported providers list
- Include PPIO logo and UI components for OAuth settings
- Support charge and billing URL redirects for PPIO
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
* fix: fix url
* fix: fix redirect url
* feat: add PPIO OAuth login
* fix: migrate
* fix: migrate
* fix: ppio migrate
---------
Co-authored-by: Claude <noreply@anthropic.com>
* fix: add custom prompt setting for translate model
Introduces a UI section in TranslateSettings to allow users to view and edit the custom prompt for the translation model. The prompt is now saved to the database and can be toggled for display in the settings modal.
* fix: add reset button for translate prompt and improve prompt editing
Introduced a reset button to restore the translate prompt to its default value. Updated the prompt editing area to use local state, improved UI with a rounded Textarea, and ensured prompt changes are dispatched to the store.
* refactor: bidirectional settings layout in TranslatePage
Removed unnecessary margin and conditional wrapper for the bidirectional settings. The Space component is now only rendered when bidirectional mode is enabled, improving layout clarity and reducing extra DOM nesting.
* Update TranslatePage.tsx
- Updated WebSearchMiddleware to utilize the new smartLinkConverter structure, allowing for better handling of buffered content and fallback logic.
- Introduced flushLinkConverterBuffer function to clear remaining buffered content at stream end.
- Modified convertLinks and smartLinkConverter functions to return structured results indicating whether content was buffered.
- Enhanced unit tests to cover new functionality and edge cases for link conversion and buffering behavior.
* test: add comprehensive tests for CopyButton component
- Add tests for basic rendering and functionality
- Add clipboard API mocking and error handling
- Add tests for custom props (size, tooltip, label)
- Add edge case testing (empty text, special characters)
- Improve component test coverage
Signed-off-by: Jason Young <farion1231@gmail.com>
* fix: resolve linting issues in CopyButton tests
- Sort imports alphabetically
- Remove trailing whitespace
- Add final newline
Signed-off-by: Jason Young <farion1231@gmail.com>
* refactor: consolidate similar test cases in CopyButton tests
- Merge 'should render copy icon' and 'should render with basic structure'
- Merge 'should apply custom size to icon' and 'should apply custom size to label'
- Reduce test duplication while maintaining full coverage
- Address maintainer feedback for better test organization
Signed-off-by: Jason Young <farion1231@gmail.com>
---------
Signed-off-by: Jason Young <farion1231@gmail.com>
fix(AddAssistantPopup): adjust body padding for improved layout consistency.
style(Messages): modify padding in ScrollContainer for better spacing; add missing line for groupedMessages.
style(Prompt): update padding and margin for improved layout aesthetics.
- Increased the default timeout value from 5 minutes to 10 minutes in constant.ts.
- Updated GeminiAPIClient and ImageGenerationMiddleware to utilize the new defaultTimeout constant for API call timeouts, ensuring consistent timeout handling across the application.
* feat(TopicsHistory): add sorting functionality for topics and update UI components
* refactor(assistants): remove console log from updateTopicUpdatedAt function
* refactor(TopicsHistory): update topic date display to use dynamic sorting type
* refactor(Messages): enhance ImageBlockGroup to dynamically adjust grid columns based on block count
* fix(ImageBlock): update maxHeight style to use responsive value for better layout
* feat(TranslatePage): replace ReactMarkdown with MarkdownIt.
* fix: line wrapping in plain text and shiki code block
---------
Co-authored-by: one <wangan.cs@gmail.com>
* feat: add country flag emoji support and enhance UI components
* Added country-flag-emoji-polyfill to package.json and yarn.lock
* Integrated polyfill in AddAgentPopup, GeneralSettings, and AssistantPromptSettings components
* Updated emoji rendering styles for better visual consistency
* fix: update country flag emoji polyfill to use 'Twemoji Country Flags'
* feat: enhance emoji components with country flag support
* Integrated country-flag-emoji-polyfill in EmojiIcon, EmojiPicker, and AssistantItem components.
* Updated font-family styles across various components for consistent emoji rendering.
* Removed redundant polyfill calls from AddAgentPopup and AssistantPromptSettings.
* refactor: streamline country flag emoji integration
* Removed redundant polyfill calls from EmojiIcon, AssistantItem, and GeneralSettings components.
* Updated EmojiPicker to use a local font file for country flag emojis.
* Added country flag font import in index.scss for improved styling consistency.
* format code
* refactor: standardize country flag font usage across components
* Introduced a new CSS class for country flag font to streamline styling.
* Updated various components (GeneralSettings, EmojiIcon, EmojiAvatar, AssistantPromptSettings, TranslatePage) to utilize the new class for consistent font application.
* Removed inline font-family styles to enhance maintainability.
* refactor: update font styles for improved consistency and maintainability
* Added Windows-specific font configuration in font.scss for better emoji rendering.
* Removed inline font-family styles from various components (EmojiAvatar, GeneralSettings, AssistantPromptSettings, TranslatePage) to enhance code clarity and maintainability.
* refactor: remove inline font-family styles from EmojiIcon for improved maintainability
- Replaced direct API call for reading binary images with FileManager's readBinaryImage method to streamline image handling in the ImageGenerationMiddleware.
* refactor: update Gemini and OpenAI API clients for improved reasoning model handling
- Replaced isGeminiReasoningModel with isSupportedThinkingTokenGeminiModel in GeminiAPIClient for better model validation.
- Enhanced OpenAIAPIClient to support additional configurations for reasoning efforts and thinking budgets based on model type.
- Introduced new thinking tags for Gemini models in ThinkingTagExtractionMiddleware.
- Updated model checks in models.ts to streamline reasoning model identification.
- Adjusted ThinkingButton component to differentiate between Gemini and Gemini Pro models based on regex checks.
* refactor(GeminiAPIClient): streamline reasoning configuration handling
- Simplified the logic for returning thinking configuration when reasoningEffort is undefined in GeminiAPIClient.
- Updated ApiService to include enableReasoning flag for API calls, enhancing control over reasoning capabilities.
* fix(OpenAIAPIClient): add support for non-flash Gemini models in reasoning configuration
- Introduced a check for non-flash models in the OpenAIAPIClient to enhance reasoning configuration handling for supported Gemini models.
- This change ensures that reasoning is correctly configured based on the model type, improving overall model validation.
- Updated model checks in isFunctionCallingModel, isEmbeddingModel, isVisionModel, and isReasoningModel functions to consider model.id for 'doubao' provider.
- Improved isOpenAIWebSearchModel to include additional conditions for model.id.
* fix(migrate): handle state return in migration process and add upgrade channel setting
* fix(migrate): move upgrade channel setting to the correct migration step
* feat(AppUpdater): add support for pre-release versions and enhance feed URL logic
- Introduced a new FeedUrl for the lowest pre-release version.
- Updated AppUpdater to handle early access and upgrade channel settings more effectively.
- Enhanced IPC logging for early access and upgrade channel changes.
- Refactored feed URL setting logic to streamline update processes.
* fix(AppUpdater, ipc): enhance early access and upgrade channel handling
- Added checks to prevent unnecessary cancellation of downloads when early access and upgrade channel settings remain unchanged.
- Updated IPC handlers to ensure early access is enabled when switching upgrade channels if it was previously disabled.
- Improved logging for better traceability of changes in early access and upgrade channel settings.
* delete code
* delete logs
* refactor(AboutSettings): enhance upgrade channel management
- Introduced logic to determine the current upgrade channel based on version.
- Refactored available test channels to use a more structured approach with tooltips and labels.
- Updated the method for retrieving available test channels to improve clarity and maintainability.
* feat(IpcChannel, ConfigManager, AppUpdater): implement test plan and channel management
- Replaced early access features with test plan and test channel options in IpcChannel and ConfigManager.
- Updated IPC handlers to manage test plan and test channel settings, including logging enhancements.
- Refactored AppUpdater to support fetching pre-release versions based on the selected test channel.
- Modified settings and localization files to reflect the new test plan functionality.
- Adjusted AboutSettings and related components to integrate test plan management and improve user experience.
* format code
* refactor(AppUpdater, AboutSettings): improve test channel logic and localization updates
- Refactored the logic in AppUpdater to enhance the handling of test channels, ensuring correct channel retrieval based on the current version.
- Updated the AboutSettings component to include useEffect for managing test channel changes and displaying appropriate warnings.
- Modified localization files for multiple languages to clarify the behavior of test version switching, aligning with the new logic.
- Adjusted regex for visionAllowedModels and DOUBAO_THINKING_MODEL_REGEX to allow for optional suffixes.
- Enhanced isFunctionCallingModel and isDoubaoThinkingAutoModel functions to check both model.id and model.name for better matching.
Reordered the ContentSearch component to render after the Messages component within the Chat page. This change likely improves the UI flow by displaying the search functionality below the chat messages.
* feat(IpcChannel): add Webview_SetSpellCheckEnabled channel and implement spell check handling for webviews
- Introduced a new IPC channel for enabling/disabling spell check in webviews.
- Updated the registerIpc function to handle spell check settings for all webviews.
- Enhanced WebviewContainer to set spell check state on DOM ready event.
- Refactored context menu setup to accommodate webview context menus.
* refactor(ContextMenu): update methods to use Electron.WebContents instead of BrowserWindow
- Changed method signatures to accept Electron.WebContents for better context handling.
- Updated internal calls to utilize the new WebContents reference for toggling dev tools and managing spell check functionality.
* refactor(WebviewContainer): clean up import order and remove unused code
- Adjusted the import order in WebviewContainer.tsx for better readability.
- Removed redundant import of useSettings to streamline the component.
* feat(WebSearch, RAG): support RAG for external websearch
* refactor(WebSearch): handle content limit in service
* refactor: update migrate
* refactor: UI, constants, types
* refactor: migrate contentLimit to cutoffLimit
* refactor: update default rag document count
* refactor: add a helper function for merging references
* refactor: reference filtering
* feat: feedback for websearch phases
* feat: support cutoff by token
* refactor: add a warning and fix the bound of cutoff limit
* fix: not pass `dimensions` if it is not set by the user
* refactor: update i18n and error message
* refactor: improve UI
* fix: cutoff unit style
* fix(Selector): update value comparison logic to use 'some' for selected values
* feat(ModelSettings): add ChevronDown icon as suffix for Select components
feat(FileStorage): enhance open dialog to handle large files by returning size without reading content
- Updated the open method to return file size for files larger than 2GB without reading their content.
- Modified return type to include an optional content field and size property for better file handling.
修复恢复备份的时候选择超过 2GB 文件报错的问题
* feat: implement early access feature toggle and update related configurations
- Replace FeedUrl with EnableEarlyAccess in IpcChannel and ConfigManager
- Update AppUpdater to handle early access updates from GitHub
- Modify settings and localization files to reflect early access functionality
- Ensure proper integration in the renderer and preload layers
* fix: enhance error handling in AppUpdater for GitHub release fetching
- Wrap the fetch call in a try-catch block to handle potential errors when retrieving the latest non-draft version from GitHub.
- Log an error message if the fetch fails and return a default feed URL.
* refactor: remove early access feature handling from AppUpdater
- Eliminate the early access feature toggle logic from the AppUpdater class.
- Adjust the feed URL setting to ensure it retrieves the latest non-draft version from GitHub when applicable.
- Clean up unnecessary user-agent header in the fetch request.
* feat(AppUpdater): enhance update feed URL logic and disable differential downloads
- Introduced a new private method to streamline feed URL setting based on early access and IP country.
- Disabled differential downloads for compatibility with GitHub and GitCode.
- Cleaned up the checkForUpdates method for better readability and maintainability.
* refactor(AppUpdater): simplify early access feed URL logic
- Consolidated the feed URL setting logic in setEnableEarlyAccess to a single line for improved readability.
- Removed redundant conditional checks while maintaining functionality for early access updates.
* refactor(AppUpdater): update feed URL structure and remove early access setting
- Modified the return structure of the latest release URL to include the channel type.
- Removed the early access setting from the IPC handler, streamlining the update process.
- Ensured the autoUpdater channel is set based on the latest release information.
* feat(UpgradeChannel): add upgrade channel management and IPC integration
- Introduced a new UpgradeChannel enum to manage different upgrade paths (latest, rc, beta).
- Updated IpcChannel to include App_SetUpgradeChannel for setting the upgrade channel.
- Enhanced ConfigManager to store and retrieve the selected upgrade channel.
- Modified AppUpdater to fetch pre-release versions based on the selected upgrade channel.
- Updated settings UI to allow users to select their preferred upgrade channel with tooltips for guidance.
- Localized new strings for upgrade channel options in multiple languages.
* refactor(AboutSettings): update version type detection and localize upgrade channel tooltips
- Changed version type detection to use the UpgradeChannel enum for better clarity.
- Localized success messages for switching upgrade channels to enhance user experience.
* chore: update version to 1.4.4-beta.1 and refactor upgrade channel handling in AboutSettings
- Updated package version to 1.4.4-beta.1.
- Renamed version type detection function to getVersionChannel for clarity.
- Refactored available version options to getAvailableTestChannels for better organization.
- Added logic to clear update info when switching upgrade channels and when toggling early access settings.
* chore: update version to 1.4.4 in package.json
* fix lint error
* feat(AppUpdater): enhance upgrade channel management and localization
- Added cancellation functionality for ongoing downloads in AppUpdater.
- Introduced a new upgrade channel option for the latest stable version.
- Updated IPC handlers to cancel downloads when changing early access settings or upgrade channels.
- Localized new strings for the latest version option in multiple languages.
- Refactored AboutSettings to include the latest version in the upgrade channel selection.
* refactor(AboutSettings): remove version channel detection logic
- Eliminated the getVersionChannel function to simplify version handling.
- Updated AboutSettings to streamline upgrade channel management.
* feat(AboutSettings): set default upgrade channel to latest
- Updated the AboutSettings component to set the default value of the upgrade channel to the latest option, enhancing user experience in channel selection.
* refactor(AboutSettings): simplify upgrade channel change handling
- Removed individual success messages for different upgrade channels in the handleUpgradeChannelChange function, streamlining the code and improving maintainability.
* refactor: file actions into FileAction service (#7413)
* refactor: file actions into FileAction service
Moved file sorting, deletion, and renaming logic from FilesPage to a new FileAction service for better modularity and reuse. Updated FileList and FilesPage to use the new service functions, and improved the delete button UI in FileList.
* fix: add tag collapse state management for assistants (#7436)
Add tag collapse state management for assistants
Introduces a collapsedTags state to manage the collapsed/expanded state of tag groups in the assistants list. Updates useTags and AssistantsTab to use this state, and adds actions to toggle and initialize tag collapse in the Redux store.
* fix(model): doubao thinking param (#7499)
* feat: Implement occupied directories handling during data copy (#7485)
* feat: Implement occupied directories handling during data copy
- Added `occupiedDirs` constant to manage directories that should not be copied.
- Enhanced the `copyOccupiedDirsInMainProcess` function to copy occupied directories to a new app data path in the main process.
- Updated IPC and preload APIs to support passing occupied directories during the copy operation.
- Modified the DataSettings component to utilize the new copy functionality with occupied directories.
* fix: Improve occupied directories handling during data copy
- Updated the filter logic in the `registerIpc` function to resolve directory paths correctly.
- Modified the `DataSettings` component to pass the correct occupied directories format during the copy operation.
* feat: add appcode (#7507)
Co-authored-by: zhaochenxue <zhaochenxue@bixin.cn>
* fix: non streamoutput sometimes (#7512)
* feat(migrate): add default settings for assistants during migration
- Introduced a new migration step to assign default settings for assistants that lack configuration.
- Default settings include temperature, context count, and other parameters to ensure consistent behavior across the application.
* chore(store): increment version number to 115 for persisted reducer
* Revert "feat: Update API Key Management Interface (#3444)"
This reverts commit 31b3ce1049.
* feat: 一些UI上的优化和重构 (#7479)
- 调整AntdProvider中主题配置，包括颜色、尺寸
- 重构聊天气泡模式的样式
- 重构多选模式的样式
- 添加Selector组件取代ant Select组件
- 重构消息搜索弹窗界面
- 重构知识库搜索弹窗界面
- 优化其他弹框UI
* fix: bailian reranker (#7518)
* feat: implement Python MCP server using existing Pyodide infrastructure (#7506)
* refactor: rename isWindows to isWin for consistency across main/renderer (#7530)
refactor: rename isWindows to isWin for consistency across components
* refactor: data migration modal logic in DataSettings (#7503)
* refactor: data migration modal logic in DataSettings
Moved showProgressModal and startMigration functions inside the useEffect hook and added t as a dependency. This improves encapsulation and ensures translation updates are handled correctly.
* remove trailing whitespace in DataSettings.tsx
Cleaned up a line by removing unnecessary trailing whitespace in the DataSettings component.
* fix: clear search cache on resending (#7510)
* fix: Resolve vllm bad request caused by always sending dimensions in embedding requests (#7525)
fix(知识库): 将dimensions字段改为可选并修复相关逻辑
* feat: Support custom registry address when configuring mcp for npm & fix lint error (#7531)
* feat: Support custom registry address when configuring mcp for npm
* fix: lint
* refactor(GeminiAPIClient): separate model and user message handling to adapt vertex (#7511)
- Introduced a new modelParts array to manage model-related messages separately from user messages.
- Updated the logic to push model messages to currentReqMessages only if they exist, improving clarity and structure.
- Adjusted the return order of messages in buildSdkMessages to ensure history is appended correctly.
- Enhanced McpToolChunkMiddleware to reset tool processing state output when output is present.
* feat: enhance WindowFooter with show/hide functionality for UI elements
- Added state management to control visibility of UI elements in the WindowFooter.
- Implemented a timer to automatically hide elements after a period of inactivity.
- Updated hotkey handlers to reset the visibility timer on user interaction.
- Modified styled component to reflect the new visibility logic.
* fix(SelectionAssistant): opacity slider too slow when sliding in settings page (#7537)
feat: enhance opacity control in Selection Assistant Settings
- Added state management for opacity value in SelectionAssistantSettings component.
- Updated Slider component to use the new opacity state instead of the previous actionWindowOpacity variable.
- Ensured onChangeComplete updates the actionWindowOpacity accordingly.
* feat(AihubmixAPIClient): add getBaseURL method to handle client base URL retrieval
* fix(migrate): restore upgradeChannel setting in migration logic
- Reintroduced the upgradeChannel setting to the state during the migration process, ensuring it defaults to LATEST when applicable.
- Adjusted the migration logic to maintain consistency in settings management.
---------
Co-authored-by: 自由的世界人 <3196812536@qq.com>
Co-authored-by: one <wangan.cs@gmail.com>
Co-authored-by: chenxue <DDU1222@users.noreply.github.com>
Co-authored-by: zhaochenxue <zhaochenxue@bixin.cn>
Co-authored-by: SuYao <sy20010504@gmail.com>
Co-authored-by: kangfenmao <kangfenmao@qq.com>
Co-authored-by: Teo <cheesen.xu@gmail.com>
Co-authored-by: Chen Tao <70054568+eeee0717@users.noreply.github.com>
Co-authored-by: LiuVaayne <10231735+vaayne@users.noreply.github.com>
Co-authored-by: fullex <106392080+0xfullex@users.noreply.github.com>
Co-authored-by: Wang Jiyuan <59059173+EurFelux@users.noreply.github.com>
Co-authored-by: 陈天寒 <silenceboychen@gmail.com>
Co-authored-by: fullex <0xfullex@gmail.com>
- Added a new condition to the Gemini reasoning model check to include models with IDs starting with 'gemini' and containing 'thinking'.
- Refactored the CitationsList component to improve the structure of popover content for web search and knowledge citations.
- Updated styled components for better layout and responsiveness in the citation popover.
- Adjusted margin styles in ErrorBlock for consistent spacing.
* feat(Messages): integrate Scrollbar component into Message and MessageGroup styled containers
* style(Messages): add margin-top to MessageFooter for improved layout
* fix(SelectionToolbar): update regex to remove background styles more accurately
refactor: update selection toolbar styles and structure
- Enhanced the selection toolbar's HTML structure for better readability.
- Updated CSS variables for improved theming and consistency across the toolbar.
- Refactored the styled components in SelectionToolbar.tsx to utilize new CSS variables for layout and styling.
- Added support for hover states and improved button styling for better user experience.
feat: enhance opacity control in Selection Assistant Settings
- Added state management for opacity value in SelectionAssistantSettings component.
- Updated Slider component to use the new opacity state instead of the previous actionWindowOpacity variable.
- Ensured onChangeComplete updates the actionWindowOpacity accordingly.
- Added state management to control visibility of UI elements in the WindowFooter.
- Implemented a timer to automatically hide elements after a period of inactivity.
- Updated hotkey handlers to reset the visibility timer on user interaction.
- Modified styled component to reflect the new visibility logic.
- Introduced a new modelParts array to manage model-related messages separately from user messages.
- Updated the logic to push model messages to currentReqMessages only if they exist, improving clarity and structure.
- Adjusted the return order of messages in buildSdkMessages to ensure history is appended correctly.
- Enhanced McpToolChunkMiddleware to reset tool processing state output when output is present.
* refactor: data migration modal logic in DataSettings
Moved showProgressModal and startMigration functions inside the useEffect hook and added t as a dependency. This improves encapsulation and ensures translation updates are handled correctly.
* remove trailing whitespace in DataSettings.tsx
Cleaned up a line by removing unnecessary trailing whitespace in the DataSettings component.
* feat(migrate): add default settings for assistants during migration
- Introduced a new migration step to assign default settings for assistants that lack configuration.
- Default settings include temperature, context count, and other parameters to ensure consistent behavior across the application.
* chore(store): increment version number to 115 for persisted reducer
* feat: Implement occupied directories handling during data copy
- Added `occupiedDirs` constant to manage directories that should not be copied.
- Enhanced the `copyOccupiedDirsInMainProcess` function to copy occupied directories to a new app data path in the main process.
- Updated IPC and preload APIs to support passing occupied directories during the copy operation.
- Modified the DataSettings component to utilize the new copy functionality with occupied directories.
* fix: Improve occupied directories handling during data copy
- Updated the filter logic in the `registerIpc` function to resolve directory paths correctly.
- Modified the `DataSettings` component to pass the correct occupied directories format during the copy operation.
Add tag collapse state management for assistants
Introduces a collapsedTags state to manage the collapsed/expanded state of tag groups in the assistants list. Updates useTags and AssistantsTab to use this state, and adds actions to toggle and initialize tag collapse in the Redux store.
* refactor: file actions into FileAction service
Moved file sorting, deletion, and renaming logic from FilesPage to a new FileAction service for better modularity and reuse. Updated FileList and FilesPage to use the new service functions, and improved the delete button UI in FileList.
* feat: enhance API key management in ApiCheckPopup: allow users to add new API key
- Enhanced ApiCheckPopup component to allow users to add new API key, including validation for duplicate entries and improved user feedback.
* feat: update localization strings and refactor API key management components
- Added "Invalid API key" message to localization files for English, Japanese, Russian, Simplified Chinese, and Traditional Chinese.
- Refactored API key management by replacing the ApiCheckPopup with a new ApiKeyList component, enhancing user experience and modularity in handling API keys across provider settings.
* refactor: update OAuthButton and ApiKeyList components for improved UI and localization
- Commented out the translation key in OAuthButton for future use.
- Removed unnecessary localization strings related to API key tips across multiple languages.
- Enhanced ApiKeyList component with styled components for better layout and user interaction.
- Updated ProviderSetting and WebSearchProviderSetting to streamline API key management UI.
* refactor: streamline ApiKeyList component and update localization strings
- Removed the "Check Multiple API Keys" translation key from English, Japanese, Russian, Simplified Chinese, and Traditional Chinese localization files.
- Updated ApiKeyList component to eliminate the model prop, enhancing its simplicity and usability.
- Improved error handling in API key validation by integrating model selection directly within the check process.
* feat: add latency tooltip to API key validation in ApiKeyList component and update localization strings
- Introduced a latency tooltip in the ApiKeyList component to display the time taken for API key validation.
- Updated localization files for English, Japanese, Russian, Simplified Chinese, and Traditional Chinese to include the new latency tooltip string.
* refactor: remove unused imports in WebSearchProviderSetting component
* refactor: improve error handling and latency tracking in ApiKeyList component
- Enhanced error handling during model selection to prevent failures when the user cancels the popup.
- Introduced latency tracking for API key validation, ensuring accurate measurement of response times.
- Streamlined the code for better readability and maintainability.
* refactor: improve styling in ApiKeyList component for better UI consistency
- Updated padding styles for error messages and list items in the ApiKeyList component to enhance visual clarity and user experience.
- Adjusted Card component properties to ensure consistent styling across the interface.
* refactor: extract key formatting logic into a separate function in ApiKeyList component
- Created a new function `formatAndConvertKeysToArray` to handle the formatting and conversion of API keys into an array of unique key objects.
- Updated the state initialization and effect hook in the ApiKeyList component to utilize the new function, improving code readability and maintainability.
* refactor: conditionally render API key section for non-copilot providers
- Updated the ProviderSetting component to conditionally display the API key section only for providers other than 'copilot', improving the user interface and experience.
- Maintained existing functionality for API key management while enhancing code readability.
* refactor: enhance ApiKeyList component for copilot provider handling
- Introduced a new condition to manage the rendering and functionality of buttons in the ApiKeyList component based on the provider type, specifically for 'copilot'.
- Updated the ProviderSetting component to ensure the API key section is consistently displayed for all providers, improving overall user experience and code clarity.
* fix model type error
* feat(ApiKeyList): exclude rerank models from being checked for API key validation after #3969 is merged
* refactor(ApiKeyList): conditionally render check and remove buttons based on key statuses
* refactor(ApiKeyList): using Promise.all for improved performance after #4066 is merged
* refactor(ProviderSettings): update API key display and tooltip integration for improved layout and accessibility
* fix(ApiKeyList): prevent notifications from showing when checking multiple API keys
* feat(ApiKeyList): enhance API key handling with improved key formatting and auto-focus logic for add button
* refactor: clean up WebSearchProviderSetting component
* refactor(ApiKeyList): replace icon buttons with styled components for save and cancel actions
* refactor: API key list UI and remove unused components
Simplified the API key list UI by removing custom styled components for status and actions, replacing them with Ant Design icons and buttons. Improved the key checking logic and removed the tooltip for key check results. Also removed an unused help text in ProviderSetting.
* refactor: add edit functionality to API key list
Introduces the ability to edit existing API keys in the ApiKeyList component. Removes custom save/cancel icon buttons in favor of standard input blur/enter and icon actions. Also adjusts styling for help text in ProviderSetting.
* refactor(ApiKeyList): enhance key status display with tooltips and color coding
* feat(i18n): add "checking" status message in multiple languages
* feat(ApiKeyList): enhance API key management with confirmation for deletion and improved state handling
- Added confirmation for deleting API keys, allowing users to confirm before removal.
- Introduced a cancel state for adding new keys to improve user experience.
- Enhanced key status updates to prevent unnecessary re-renders.
- Improved UI interactions with better handling of edit and cancel actions.
- Added escape key functionality for canceling edits and new key entries.
* fix(ApiKeyList): adjust layout of API key list for improved spacing and alignment
- Updated the Flex component to justify content between elements, enhancing the visual layout of the API key list.
- Minor style adjustment to maintain consistency in the user interface.
* fix(ApiKeyList): refine padding for API key list items to enhance visual consistency
- Adjusted padding for API key list text and items to improve overall layout and alignment.
- Ensured consistent spacing across different states of the API key list.
---------
Co-authored-by: Pleasurecruise <3196812536@qq.com>
Co-authored-by: suyao <sy20010504@gmail.com>
* fix: display updated timestamp when available in knowledge base
- Add updated_at field when creating knowledge items
- Show updated_at timestamp if it's newer than created_at
- Fallback to created_at if updated_at is not available or older
Fixes#4587
Signed-off-by: Ying-xi <62348590+Ying-xi@users.noreply.github.com>
* refactor(knowledge): extract display time logic into a reusable function
Signed-off-by: Ying-xi <62348590+Ying-xi@users.noreply.github.com>
---------
Signed-off-by: Ying-xi <62348590+Ying-xi@users.noreply.github.com>
* feat(ContextMenu): add spell check and dictionary suggestions to context menu
- Implemented spell check functionality in the context menu with options to learn spelling and view dictionary suggestions.
- Updated WindowService to enable spellcheck in the webview.
- Enabled spell check in Inputbar and MessageEditor components.
* feat(SpellCheck): implement spell check language settings and initialization
- Added support for configuring spell check languages based on user-selected language.
- Introduced IPC channel for setting spell check languages.
- Updated settings to manage spell check enablement and languages.
- Enhanced UI to allow users to toggle spell check functionality and select languages.
- Default spell check languages are set based on the current UI language if none are specified.
* refactor(SpellCheck): enhance spell check language mapping and UI settings
- Updated spell check language mapping to default to English for unsupported languages.
- Improved UI logic to only update spell check languages when enabled and no manual selections are made.
- Added a new selection component for users to choose from commonly supported spell check languages.
* feat(SpellCheck): integrate spell check functionality into Inputbar and MessageEditor
- Added enableSpellCheck setting to control spell check functionality in both Inputbar and MessageEditor components.
- Updated spellCheck prop to utilize the new setting, enhancing user experience by allowing customization of spell check behavior.
* refactor(SpellCheck): move spell check initialization to WindowService
- Removed spell check language initialization from index.ts and integrated it into WindowService.
- Added setupSpellCheck method to configure spell check languages based on user settings.
- Enhanced error handling for spell check language setup.
* feat(SpellCheck): add enable spell check functionality and IPC channel
- Introduced a new IPC channel for enabling/disabling spell check functionality.
- Updated the preload API to include a method for setting spell check enablement.
- Modified the main IPC handler to manage spell check settings based on user input.
- Simplified spell check language handling in the settings component by directly invoking the new API method.
* refactor(SpellCheck): remove spellcheck option from WindowService configuration
- Removed the spellcheck property from the WindowService configuration object.
- This change streamlines the configuration setup as spell check functionality is now managed through IPC channels.
* feat(i18n): add spell check translations for Japanese, Russian, and Traditional Chinese
- Added new translations for spell check functionality in ja-jp, ru-ru, and zh-tw locale files.
- Included descriptions and language selection options for spell check settings to enhance user experience.
* feat(migrate): add spell check configuration migration
- Implemented migration for spell check settings, disabling spell check and clearing selected languages in the new configuration.
- Enhanced error handling to ensure state consistency during migration process.
* fix(migrate): ensure spell check settings are updated safely
- Added a check to ensure state.settings exists before modifying spell check settings during migration.
- Removed redundant error handling that returned the state unmodified in case of an error.
* fix(WindowService): set default values for spell check configuration and update related UI texts
* refactor(Inputbar, MessageEditor): remove contextMenu attribute and add context menu handling in MessageEditor
---------
Co-authored-by: beyondkmp <beyondkmkp@gmail.com>
* fix: update localization files for data migration warnings and path validation messages
* fix: update app data path validation and localization messages for installation path consistency
* fix: enhance app data flushing process by adding connection closure and delay in DataSettings component
* Revert "refactor(middleware): Add error property to CompletionResult and handle errors when checking API (#7407)"
This reverts commit 50d6f1f831.
* fix: use shouldThrow param in checkApi
* fix: move initAppDataDir function inline and remove export from utils/file.ts
* fix some bugs
* fix shouldcopy error
* fix: handle appDataPath initialization and update logic in file.ts; update defaultChecked in DataSettings component
* fix: improve appDataPath handling and migration logic in file.ts
* fix: add error message for selecting the same app data path in DataSettings component and update localization files
* fix: ensure migration confirmation modal is shown correctly in DataSettings component
* feat: add new IPC channel for retrieving data path from arguments and update related components for migration handling
* fix: update app data path validation to check for prefix match in DataSettings component
* refactor: simplify data migration logic in DataSettings component by removing unnecessary flag
* fix: update initAppDataDir invocation to check for app packaging status in bootstrap.ts
* chore(package): add node-stream-zip for zip file extraction in install-bun script
* refactor(install-uv): replace AdmZip with node-stream-zip for improved zip file extraction
* fix(install-uv): ensure correct extraction of uv binary for Unix/Linux/macOS
* refactor(install-uv): remove redundant file handling and cleanup for Unix/Linux/macOS installation
* fix(install-uv): update tar extraction command to strip leading components for Unix/Linux/macOS
* fix(install-uv): clarify comment for zip file extraction on Windows
* fix(install-bun): correct extraction directory for bun binary
* fix(install-bun, install-uv): update default versions and improve zip extraction process
* fix(install-bun): remove redundant cleanup of source directory after bun installation
- Removed unused MCP server handling from Inputbar and MessagesService.
- Updated ApiService to fetch active MCP servers directly from the store.
- Deprecated knowledgeBaseIds and enabledMCPs in Message types and related functions.
- Cleaned up related utility functions to enhance code clarity and maintainability.
* fix: openai response tool use
- Added OpenAIResponseStreamListener interface for handling OpenAI response streams.
- Implemented attachRawStreamListener method in OpenAIResponseAPIClient to manage raw output.
- Updated RawStreamListenerMiddleware to integrate OpenAI response handling.
- Refactored BaseApiClient to remove unused attachRawStreamListener method.
- Improved buildSdkMessages to handle OpenAI response formats.
* fix: remove logging from StreamAdapterMiddleware
- Removed Logger.info call from StreamAdapterMiddleware to streamline output and reduce unnecessary logging.
* fix: update attachRawStreamListener to return a Promise
- Changed attachRawStreamListener method in OpenAIResponseAPIClient to be asynchronous, returning a Promise for better handling of raw output.
- Updated RawStreamListenerMiddleware to await the result of attachRawStreamListener, ensuring proper flow of data handling.
* refactor: enhance attachRawStreamListener to return a ReadableStream
- Updated the attachRawStreamListener method in OpenAIResponseAPIClient to return a ReadableStream, allowing for more efficient handling of streamed responses.
- Modified RawStreamListenerMiddleware to accommodate the new return type, ensuring proper integration of the transformed stream into the middleware flow.
* refactor: update getResponseChunkTransformer to accept CompletionsContext
- Modified the getResponseChunkTransformer method in BaseApiClient and its implementations to accept a CompletionsContext parameter, enhancing the flexibility of response handling.
- Adjusted related middleware and client classes to ensure compatibility with the new method signature, improving the overall integration of response transformations.
* refactor: update getResponseChunkTransformer to accept CompletionsContext
- Modified the getResponseChunkTransformer method in AihubmixAPIClient to accept a CompletionsContext parameter, enhancing the flexibility of response handling.
- Ensured compatibility with the updated method signature across related client classes.
- Added error handling for tool fetching to log errors and return an empty array if a server fails to respond.
- Changed from Promise.all to Promise.allSettled to ensure all tool fetching attempts are accounted for, filtering out any rejected promises.
- Added debug logging in buildSdkMessages for better traceability.
- Improved handling of tool calls in the transform method to correctly index multiple tool uses.
- Enqueued additional response types to enhance the output structure for better integration with the streaming API.
- Refactored event listener attachment for clarity and maintainability.
* fix: series bugs of quick assistant
* fix: update quick assistant ID handling and improve error management in HomeWindow
* refactor(HomeWindow, Messages): streamline clipboard handling and improve component structure
- Removed unused imports and hotkey functionality from Messages component.
- Refactored clipboard management in HomeWindow to use refs for better performance.
- Enhanced user input handling and state management in HomeWindow.
- Updated InputBar to accept assistant prop instead of model for better clarity.
- Improved Footer component to handle copy functionality and pin state more effectively.
* Enhance Footer component: add rotation animation to pin icon and adjust margin
- Updated the Pin icon in the Footer component to include a rotation animation based on the pin state.
- Adjusted the margin of the PinButtonArea for improved layout consistency.
* refactor(HomeWindow): improve clipboard handling and input placeholder logic
- Updated clipboard reading logic to check for document focus in addition to startup settings.
- Consolidated key event handling to streamline input processing.
- Enhanced placeholder logic in InputBar to reflect the current assistant's name or model more accurately.
* fix: initialize messageContents and improve message handling in GeminiAPIClient; add new Gemini model to configuration
* refactor: streamline message handling in GeminiAPIClient; enhance message extraction from SDK payload
* feat: Add app data path selection and relaunch functionality
* Introduced new IPC channels for selecting and setting the app data path.
* Implemented logic to initialize the app data path on startup.
* Added confirmation modal for changing the app data directory in the settings.
* Updated translations for new app data path features in multiple languages.
* feat: Implement user data copying and app data path management
* Added IPC channels for copying user data to a new location and setting the app data path.
* Enhanced the user interface to support data copying with progress notifications.
* Updated translations to reflect new features related to app data management.
* Refactored file utility functions to streamline data path handling.
* refactor: update IPC channel names and streamline app data path handling
- Renamed IPC channels for selecting app data path and copying user data for clarity.
- Simplified the logic for selecting and setting app data paths, removing unnecessary success/error handling.
- Updated related functions and components to reflect the new IPC channel names and improved data handling.
- Removed unused copyUserDataToNewLocation function to clean up the codebase.
* fix: update app data directory selection text in multiple locales
- Changed the text for selecting the app data directory from "Select Directory" to "Modify Directory" in English, Japanese, Russian, Simplified Chinese, and Traditional Chinese locales to better reflect the action being performed.
* refactor: remove redundant success messages in DataSettings component
- Eliminated unnecessary success messages related to app data copying and app relaunching to streamline user feedback and improve code clarity.
* refactor: streamline file utility functions and update app data initialization
- Moved `getDataPath` function to the `utils/index.ts` for better organization and accessibility.
- Renamed `initUserDataDir` to `initAppDataDir` for clarity in its purpose.
- Removed commented-out code in `ConfigManager` to enhance code cleanliness.
* refactor: update import paths and localization strings for app data
- Refactored import statements for `getDataPath` to streamline utility access.
- Updated localization strings for app data in English, Japanese, Russian, Simplified Chinese, and Traditional Chinese to enhance clarity and consistency.
* update i18n
* add fc
* fix: handle errors in app data path retrieval
- Added error handling to the `getAppDataPathFromConfig` function to return null if the configuration file cannot be read or parsed, improving robustness.
* refactor: simplify app data path handling in IPC
- Removed error handling for setting the app data path in the IPC channel, streamlining the process by directly updating the configuration and user data path without try-catch blocks.
* fix: update userData path handling for portable applications
- Modified the initAppDataDir function to set the userData path based on the PORTABLE_EXECUTABLE_DIR environment variable, ensuring compatibility with portable application setups.
* feat: enhance app data path migration with progress indication
- Implemented a loading modal with progress tracking during the app data path migration process.
- Added visual feedback using a progress bar to inform users of the copying status.
- Improved error handling and user notifications for successful and failed migrations.
- Refactored the modal confirmation logic to streamline user interactions during the path selection and migration process.
* feat: add migration paths and update UI for data migration process
- Introduced new translation keys for migration paths in Japanese, Russian, Simplified Chinese, and Traditional Chinese.
- Enhanced the DataSettings component with a structured layout for displaying original and new paths during data migration.
- Updated modal titles and content to improve user experience during the migration process.
* feat: enhance data migration process with improved UI and progress tracking
- Refactored the DataSettings component to streamline the data migration workflow.
- Added a new function to display progress during the migration process, enhancing user feedback.
- Updated modal logic to improve clarity and user experience when selecting new app data paths.
- Implemented error handling and notifications for successful and failed migrations.
* feat: add stop quit app functionality during data migration
- Introduced a new IPC channel to manage the application's quit behavior during data transfer.
- Updated the DataSettings component to prevent the app from quitting while migration is in progress, enhancing user experience.
- Improved modal configurations for better responsiveness and visual appeal.
* feat: enhance app data path handling and localization updates
- Updated IPC handler to use 'filePath' for clarity in app data path management.
- Improved validation to ensure the new app data path is not the root path, enhancing user feedback during path selection.
- Added new translation keys for error messages related to app data path selection in English, Japanese, Russian, Simplified Chinese, and Traditional Chinese, improving localization support.
* feat: add write permission check and enhance quit prevention during data migration
- Introduced a new IPC channel to check write permissions for the app data path.
- Updated the DataSettings component to validate write permissions before proceeding with data migration.
- Enhanced the quit prevention logic to include a reason for blocking the app from quitting during data transfer.
- Added new localization keys for error messages related to write permissions in multiple languages, improving user feedback.
* feat: enhance confirmation modal in DataSettings component
- Updated the confirmation modal to include danger styling for the OK button, improving visual feedback.
- Added localized text for the OK and Cancel buttons, enhancing user experience through better accessibility.
* feat: add localization keys and improve quit prevention during data migration
- Added new localization keys for data migration, including titles and original path labels, enhancing user experience.
- Updated the DataSettings component to ensure the app does not quit during data migration, improving reliability and user feedback.
* feat(DataSettings): add data copying option and update related messages
- Introduced a switch to allow users to choose whether to copy data from the original directory when changing the app data path.
- Updated user notifications and progress messages to reflect the new functionality, including warnings about data copying.
- Enhanced localization files for multiple languages to include new strings related to data copying options and notifications.
* fix(DataSettings): set default for data copying option to true
- Updated the DataSettings component to set the default state of the data copying option to true.
- Added a new CopyDataContent component to enhance the user interface by displaying the data copying option alongside the existing path settings.
- Improved layout by ensuring proper spacing and alignment for better user experience.
---------
Co-authored-by: beyondkmp <beyondkmkp@gmail.com>
* feat: Add pricing configuration and display for models
- Introduce model pricing fields in ModelEditContent
- Add price calculation and display in MessageTokens
- Update localization files with price-related translations
- Extend Model type with optional pricing information
* fix: Correct currency symbol placement in message token pricing display
* feat: Add custom currency support in model pricing configuration
- Introduce custom currency option in ModelEditContent
- Update localization files with custom currency translations
- Enhance currency symbol selection with custom input
- Improve input styling for pricing configuration
* fix(OpenAIProvider): ensure messages.content of the request is string
* Update ModelEditContent.tsx
* fix(model-price): remove duplicate button
* fix: build error
---------
Co-authored-by: 自由的世界人 <3196812536@qq.com>
* WIP
* feat: integrate Vertex AI support and enhance service account configuration
- Added Vertex AI service integration with authentication via service accounts.
- Implemented IPC channels for Vertex AI authentication and cache management.
- Updated UI components to support service account configuration, including private key and client email fields.
- Enhanced localization for Vertex AI settings in multiple languages.
- Refactored AiProvider to support dynamic provider creation for Vertex AI.
- Updated Redux store to manage Vertex AI settings and service account information.
* chore: remove debug script from package.json and clean up console log in main process
* fix: ensure async handling in useKnowledge hook for base parameters
- Updated the useKnowledge hook to await the result of getKnowledgeBaseParams when removing items, ensuring proper asynchronous behavior.
* fix: ensure async handling in KnowledgeQueue for base parameters
* fix(i18n): add English prompt placeholder to Russian localization
* chore(yarn): update yarn.lock and patch for @google/genai
* fix(AihubmixPage): update AI provider instantiation to use async create method
* refactor: update VertexAPIClient import and class definition
- Changed import statement for VertexAPIClient to use named import.
- Updated VertexProvider class to VertexAPIClient for consistency with naming conventions.
* refactor: update AiProvider instantiation across components
- Replaced the use of AiProvider.create() with the new AiProvider() constructor in AddKnowledgePopup, AihubmixPage, SiliconPage, and KnowledgeService for consistency and improved clarity.
* refactor: simplify getKnowledgeBaseParams and update API key checks
- Changed getKnowledgeBaseParams to a synchronous function for improved performance.
- Updated API key validation logic to remove unnecessary checks for 'vertexai' provider type across multiple functions.
* feat: add Cephalon provider configuration with API and website links
- Introduced a new provider configuration for Cephalon, including API URL and various website links for official resources, API key, documentation, and models.
* refactor: streamline API call in AddKnowledgePopup component
- Removed unnecessary await from the create API call in the AddKnowledgePopup component, improving code clarity and performance.
* refactor: remove unnecessary await from getKnowledgeBaseParams call
- Simplified the searchKnowledgeBase function by removing the await from getKnowledgeBaseParams, enhancing performance and code clarity.
* refactor: remove externalLiveBindings option from Rollup output configuration in electron.vite.config.ts
* fix: downgrade version in package.json and enhance AppUpdater with IP country detection
- Downgraded the application version from 1.4.2 to 1.4.1 in package.json.
- Added a new private method `_getIpCountry` in AppUpdater to fetch the user's IP country with a timeout mechanism.
- Updated the `setAutoUpdate` method to adjust the feed URL based on the detected country, improving update handling for users outside of China.
* fix: adjust timeout duration and enhance IP country logging in AppUpdater
* fix: extend timeout duration in AppUpdater for improved fetch reliability
---------
Co-authored-by: beyondkmp <beyondkmkp@gmail.com>
* feat: add quick assistant settings panel and management functionality
- Create QuickAssistantSettings component for UI
- Extend useAssistant hook with quick assistant controls
- Add settings button in ModelSettings page
- Implement temperature, context count, max tokens, and other parameters
- Connect settings to store via updateQuickAssistant action
Separate quick assistant preferences from default assistant settings for better customization.
* refactor(QuickAssistantSettings): remove maxTokens and refine UI layout
- Removed maxTokens related state, logic, and UI elements
- Simplified settings page by eliminating unused configuration
- Adjusted layout for Slider and InputNumber for better usability
- Removed fixed width from Modal to enable responsive behavior
* refactor(HomeWindow): optimize message building logic
- Removed redundant quickAssistant fetching logic
- Use `useQuickAssistant` hook directly for cleaner code
- Simplified message content concatenation method
* style(QuickAssistantSettings): Adjust spacing in settings page layout
Change the column width of sliders and input fields from 20/4 to 21/3 for a more reasonable layout
Also set the popup width to 800px to improve user experience
* feat(Quick Assistant): Add option to select assistant or model, and optimize Quick Assistant logic
- Added functionality to choose between using models or referencing other assistants
- Optimized model selection logic to automatically select based on settings
- Added relevant internationalization texts
* fix(HomeWindow): Dynamically display input box placeholder text based on quick assistant states
* refactor(QuickAssistant): remove the implement of the quick assistant feature and restructure related logic
- Remove code related to the quick assistant feature, including the useQuickAssistant hook, QuickAssistantSettings component, and associated store logic.
- Restructure the HomeWindow component to use default or specified assistants instead of the quick assistant functionality, simplifying the code structure.
* refactor(QuickAssistant): Remove custom default model for quick assistant and switch to default assistant
- Refactor quick assistant functionality, remove independent model settings, change to select via assistant ID
- Update multilingual translation text to match new features
* refactor(QuickAssistant): Remove quick assistant-related states and simplify logic
- Remove unused quick assistant states and toggle functionality, simplifying related logic
- Update multilingual files to match the new default model and assistant labels
* refactor(i18n): Unify translation keys for input field placeholders
Unify the placeholder translation keys from `model_empty` and `assistant_empty` into empty across different scenarios, streamlining code logic
* refactor(settings): simplify quick helper selection logic by directly using the preset helper
- Removed redundant helper filtering logic, directly using the preset helper as the quick helper
refactor(SelectionService): enhance logging and adjust window behavior for Windows compatibility
- Updated logInfo method to include a forceShow parameter for improved logging control.
- Ensured toolbar window is set to always on top when shown.
- Commented out setOpacity calls to prevent transparency issues on Windows 10.
* feat: use variables in topic naming
* feat: use structured conversation string
* feat: add i18n
* feat: add i18n
* feat: implement summaries for other providers
* fix: adjust new version
* feat: Structure the conversation as a JSON string when naming the topic
* fix: improve logic
* fix: improve prompts
* update fetchMessageSummary
- Updated the resolution and checksum for the app-builder-lib patch in yarn.lock.
- Modified macPackager.js and updateInfoBuilder.js to correctly reference LSMinimumSystemVersion.
- Enhanced ArchiveTarget.js and NsisTarget.js to include minimumSystemVersion in updateInfo if specified.
- Changed the fetch depth to 0 in the release workflow to ensure all history is available for tagging. This adjustment improves the accuracy of the release process.
- Modified the Rollup options to disable code splitting and enable inline dynamic imports, ensuring a single file output for the build process. This change optimizes the packaging of the Electron application.
* feat: enable minification in build configurations and clean up Windows license files
- Added minification option to the build configurations in electron.vite.config.ts to optimize output size.
- Updated after-pack.js to remove unnecessary license files on Windows, improving the packaging process.
* refactor: remove minification from build configurations in electron.vite.config.ts
- Eliminated the minification option from the build settings in electron.vite.config.ts to streamline the build process.
- This change may improve build times and simplify configuration management.
---------
Co-authored-by: beyondkmp <beyondkmkp@gmail.com>
* refactor(ImageBlock): enhance loading state presentation and improve layout responsiveness
- Wrapped the loading spinner in a new SpinnerWrapper for better alignment and presentation during streaming and processing states.
- Updated the ImageBlockGroup to use `repeat(auto-fit, minmax(...))` for more flexible grid layout, improving responsiveness across different screen sizes.
These changes enhance the user experience by providing a clearer loading indication and a more adaptable layout for image blocks.
* style(ImageBlockGroup): comment out child styling for future adjustments
- Commented out the child styling rules in ImageBlockGroup to allow for potential layout modifications without removing the code entirely.
- This change prepares the component for further enhancements while maintaining existing functionality.
* refactor(ImageBlock): replace loading spinner with Ant Design Skeleton component
- Updated the loading state presentation in ImageBlock by replacing the custom spinner with Ant Design's Skeleton component for a more consistent UI experience.
- Removed the SpinnerWrapper and simplified the return statement for better readability.
- This change enhances the visual feedback during image loading while maintaining the component's functionality.
---------
Co-authored-by: lizhixuan <zhixuan.li@banosuperapp.com>
* fix(TopicRenaming): captured activeTopic.id is outdated and causes accidental topic changing after renaming
* fix: prevent topic changing on auto renaming
* fix: filter out main text on summarizing
- Added support for setting the installation directory for the autoUpdater on Windows using NsisUpdater.
- Imported the 'path' module to dynamically determine the installation path based on the executable location.
- This change improves the updater's functionality and ensures a smoother installation experience for Windows users.
Co-authored-by: beyondkmp <beyondkmkp@gmail.com>
* chore: update jsdom dependency to patch version 26.1.0
- Changed jsdom version from ^26.0.0 to a patched version 26.1.0 in package.json and yarn.lock.
- Applied a specific patch to address issues with the jsdom package.
* chore: update package.json dependencies
- Removed outdated dependencies and added new ones to improve project functionality.
- Updated versions for several packages, including @strongtz/win32-arm64-msvc, os-proxy-config, and selection-hook.
- Reorganized dependencies and devDependencies for better clarity and maintenance.
* chore: update package dependencies and remove jsdom patch
- Replaced @cherrystudio/embedjs-libsql with @libsql/client and added @libsql/win32-x64-msvc and jsdom as new dependencies.
- Updated turndown version and removed the jsdom patch from the project.
- Ensured consistency in dependency versions across package.json and yarn.lock.
* feat: Support reasoning control for Doubao models.
* feat: Enhance model handling and support for Doubao and Gemini in API clients
- Added support for Doubao thinking modes in OpenAIAPIClient and GeminiAPIClient.
- Introduced GEMINI_FLASH_MODEL_REGEX for model identification.
- Updated models.ts to include new Doubao and Gemini model regex patterns.
- Added new image asset for ChatGPT in models.
- Enhanced reasoning control and token budget handling for Doubao models.
- Improved the Inputbar's ThinkingButton component to accommodate new thinking options.
---------
Co-authored-by: suyao <sy20010504@gmail.com>
* feat: add middleware support for OpenAIProvider with logging capabilities
- Introduced middleware functionality in OpenAIProvider to enhance completions processing.
- Created AiProviderMiddlewareTypes for defining middleware interfaces and contexts.
- Implemented sampleLoggingMiddleware for logging message content and processing times.
- Updated OpenAIProvider constructor to accept middleware as an optional parameter.
- Refactored completions method to utilize middleware for improved extensibility and logging.
* refactor: streamline OpenAIProvider initialization and middleware application
- Removed optional middleware parameter from OpenAIProvider constructor for simplicity.
- Refactored ProviderFactory to create instances of providers and apply logging middleware consistently.
- Enhanced completions method visibility by changing it from private to public.
- Cleaned up unused code related to middleware handling in OpenAIProvider.
* feat: enhance AiProvider with new middleware capabilities and completion context
- Added public getter for provider info in BaseProvider.
- Introduced finalizeSdkRequestParams hook for middleware to modify SDK-specific request parameters.
- Refactored completions method in OpenAIProvider to accept a context object, improving middleware integration.
- Updated middleware types to include new context structure and callback functions for better extensibility.
- Enhanced logging middleware to utilize new context structure for improved logging capabilities.
* refactor: enhance middleware structure and context handling in AiProvider
- Updated BaseProvider and AiProvider to utilize AiProviderMiddlewareCompletionsContext for completions method.
- Introduced new utility functions for middleware context creation and execution.
- Refactored middleware application logic to improve extensibility and maintainability.
- Replaced sampleLoggingMiddleware with a more robust LoggingMiddleware implementation.
- Added new context management features for better middleware integration.
* refactor: update AiProvider and middleware structure for improved completions handling
- Refactored BaseProvider and AiProvider to change completions method signature from context to params.
- Removed unused AiProviderMiddlewareCompletionsContext and related code for cleaner implementation.
- Enhanced middleware configuration by introducing a dedicated middleware registration file.
- Implemented logging middleware for completions to improve observability during processing.
- Streamlined middleware application logic in ProviderFactory for better maintainability.
* docs: 添加中间件编写指南文档
- 新增《如何为 AI Provider 编写中间件》文档，详细介绍中间件架构、类型及编写示例。
- 说明了中间件的执行顺序、注册方法及最佳实践，旨在帮助开发者有效创建和维护中间件。
* refactor: update completions method signatures and introduce CompletionsResult type
- Changed the completions method signature in BaseProvider and AiProvider to return CompletionsResult instead of void.
- Added CompletionsResult type definition to encapsulate streaming and usage metrics.
- Updated middleware and related components to handle the new CompletionsResult structure, ensuring compatibility with existing functionality.
- Introduced new middleware for stream adaptation to enhance chunk processing during completions.
* refactor: enhance AiProvider middleware and streaming handling
- Updated CompletionsResult type to support both OpenAI SDK stream and ReadableStream.
- Modified CompletionsMiddleware to return CompletionsResult, improving type safety.
- Introduced StreamAdapterMiddleware to adapt OpenAI SDK streams to application-specific chunk streams.
- Enhanced logging in CompletionsLoggingMiddleware to capture and return results from next middleware calls.
* refactor: update AiProvider and middleware for OpenAI completions handling
- Renamed CompletionsResult to CompletionsOpenAIResult for clarity and updated its structure to support both OpenAI SDK and application-specific streams.
- Modified completions method signatures in AiProvider and OpenAIProvider to return CompletionsOpenAIResult.
- Enhanced middleware to process and adapt OpenAI SDK streams into standard chunk formats, improving overall streaming handling.
- Introduced new middleware components: FinalChunkConsumerAndNotifierMiddleware and OpenAISDKChunkToStandardChunkMiddleware for better chunk processing and logging.
* 删除 ExtractReasoningCompletionsMiddleware.ts 文件，清理未使用的中间件代码以提高代码整洁性和可维护性。
* refactor: consolidate middleware types and improve imports
- Replaced references to AiProviderMiddlewareTypes with the new middlewareTypes file across various middleware components for better organization.
- Introduced TextChunkMiddleware to enhance chunk processing from OpenAI SDK streams.
- Cleaned up imports in multiple files to reflect the new structure, improving code clarity and maintainability.
* feat: enhance abort handling with AbortController in middleware chain
- Update CompletionsOpenAIResult interface to use AbortController instead of AbortSignal
- Modify OpenAIProvider to pass abortController in completions method return
- Update AbortHandlerMiddleware to use controller from upstream result
- Improve abort handling flexibility by exposing full controller capabilities
- Enable middleware to actively control abort operations beyond passive monitoring
This change provides better control over request cancellation and enables
more sophisticated abort handling patterns in the middleware pipeline.
* refactor: enhance AiProvider and middleware for improved completions handling
- Updated BaseProvider to expose additional methods and properties, including getMessageParam and createAbortController.
- Modified OpenAIProvider to streamline completions processing and integrate new middleware for tool handling.
- Introduced TransformParamsBeforeCompletions middleware to standardize parameter transformation before completions.
- Added McpToolChunkMiddleware for managing tool calls within the completions stream.
- Enhanced middleware types to support new functionalities and improve overall structure.
These changes improve the flexibility and maintainability of the AiProvider and its middleware, facilitating better handling of OpenAI completions and tool interactions.
* refactor: enhance middleware for recursive handling and internal state management
- Introduced internal state management in middleware to support recursive calls, including enhanced dispatch functionality.
- Updated middleware types to include new internal fields for managing recursion depth and call status.
- Improved logging for better traceability of recursive calls and state transitions.
- Adjusted various middleware components to utilize the new internal state, ensuring consistent behavior during recursive processing.
These changes enhance the middleware's ability to handle complex scenarios involving recursive calls, improving overall robustness and maintainability.
* fix(OpenAIProvider): return empty object for missing sdkParams in completions handling
- Updated OpenAIProvider to return an empty object instead of undefined when sdkParams are not found, ensuring consistent return types.
- Enhanced TransformParamsBeforeCompletions middleware to include a flag for built-in web search functionality based on assistant settings.
* refactor(OpenAIProvider): enhance completions handling and middleware integration
- Updated the completions method in OpenAIProvider to include an onChunk callback for improved streaming support.
- Enabled the ThinkChunkMiddleware in the middleware registration for better handling of reasoning content.
- Increased the maximum recursion depth in McpToolChunkMiddleware to prevent infinite loops.
- Refined TextChunkMiddleware to directly enqueue chunks without unnecessary type checks.
- Improved the ThinkChunkMiddleware to better manage reasoning tags and streamline chunk processing.
These changes enhance the overall functionality and robustness of the AI provider and middleware components.
* feat(WebSearchMiddleware): add web search handling and integration
- Introduced WebSearchMiddleware to process various web search results, including annotations and citations, and generate LLM_WEB_SEARCH_COMPLETE chunks.
- Enhanced TextChunkMiddleware to support link conversion based on the model and assistant settings, improving the handling of TEXT_DELTA chunks.
- Updated middleware registration to include WebSearchMiddleware for comprehensive search result processing.
These changes enhance the AI provider's capabilities in handling web search functionalities and improve the overall middleware architecture.
* fix(middleware): improve optional chaining for chunk processing
- Updated McpToolChunkMiddleware and ThinkChunkMiddleware to use optional chaining for accessing choices, enhancing robustness against undefined values.
- Removed commented-out code in ThinkChunkMiddleware to streamline the chunk handling process.
These changes improve the reliability of middleware when processing OpenAI API responses.
* feat(middleware): enhance AbortHandlerMiddleware with recursion handling
- Added logic to detect and handle recursive calls, preventing unnecessary creation of AbortControllers.
- Improved logging for better visibility into middleware operations, including recursion depth and cleanup processes.
- Streamlined cleanup process for non-stream responses to ensure resources are released promptly.
These changes enhance the robustness and efficiency of the AbortHandlerMiddleware in managing API requests.
* docs(middleware): 迁移步骤
* feat(middleware): implement FinalChunkConsumerMiddleware for usage and metrics accumulation
- Introduced FinalChunkConsumerMiddleware to replace the deprecated FinalChunkConsumerAndNotifierMiddleware.
- This new middleware accumulates usage and metrics data from OpenAI API responses, enhancing tracking capabilities.
- Updated middleware registration to utilize the new FinalChunkConsumerMiddleware, ensuring proper integration.
- Added support for handling recursive calls and improved logging for better debugging and monitoring.
These changes enhance the middleware's ability to manage and report usage metrics effectively during API interactions.
* refactor(migrate): update API request and response structures to TypeScript types
- Changed the definitions of `CoreCompletionsRequest` and `Chunk` to use TypeScript types instead of Zod Schemas for better type safety and clarity.
- Updated middleware and service classes to handle the new `Chunk` type, ensuring compatibility with the revised API client structure.
- Enhanced the response processing logic to standardize the handling of raw SDK chunks into application-level `Chunk` objects.
- Adjusted middleware to consume the new `Chunk` type, streamlining the overall architecture and improving maintainability.
These changes facilitate a more robust and type-safe integration with AI provider APIs.
* feat(AiProvider): implement API client architecture
- Introduced ApiClientFactory for creating instances of API clients based on provider configuration.
- Added BaseApiClient as an abstract class to provide common functionality for specific client implementations.
- Implemented OpenAIApiClient for OpenAI and Azure OpenAI, including request and response handling.
- Defined types and interfaces for API client operations, enhancing type safety and clarity.
- Established middleware schemas for standardized request processing across AI providers.
These changes lay the groundwork for a modular and extensible API client architecture, improving the integration of various AI providers.
* refactor(StreamAdapterMiddleware): simplify stream adaptation logic
- Updated StreamAdapterMiddleware to directly use AsyncIterable instead of wrapping it with rawSdkChunkAdapter, streamlining the adaptation process.
- Modified asyncGeneratorToReadableStream to accept AsyncIterable, enhancing its flexibility and usability.
These changes improve the efficiency of stream handling in the middleware.
* refactor(AiProvider): simplify ResponseChunkTransformer interface and streamline OpenAIApiClient response handling
- Changed ResponseChunkTransformer from an interface to a type for improved clarity and simplicity.
- Refactored OpenAIApiClient to streamline the response transformation logic, reducing unnecessary complexity in handling tool calls and reasoning content.
- Enhanced type safety by ensuring consistent handling of optional properties in response processing.
These changes improve the maintainability and readability of the codebase while ensuring robust response handling in the API client.
* doc(technicalArchitecture): add comprehensive documentation for AI Provider architecture
* feat(architecture): introduce AI Core Design documentation and middleware specification
- Added a comprehensive technical architecture document for the new AI Provider (`aiCore`), outlining core design principles, component details, and execution flow.
- Established a middleware specification document to define the design, implementation, and usage of middleware within the `aiCore` module, promoting a flexible and maintainable system.
- These additions provide clarity and guidance for future development and integration of AI functionalities within Cherry Studio.
* refactor(middleware): consolidate and enhance middleware architecture
- Removed deprecated extractReasoningMiddleware and integrated its functionality into existing middleware.
- Streamlined middleware registration and improved type definitions for better clarity and maintainability.
- Introduced new middleware components for handling chunk processing, web search, and reasoning tags, enhancing overall functionality.
- Updated various middleware to utilize the new structures and improve logging for better debugging.
These changes enhance the middleware's efficiency and maintainability, providing a more robust framework for API interactions.
* refactor(AiProvider): enhance API client and middleware integration
- Updated ApiClientFactory to include new SDK types for improved type safety and clarity.
- Refactored BaseApiClient to support additional parameters in the completions method, enhancing flexibility for processing states.
- Streamlined OpenAIApiClient to better handle tool calls and responses, including the introduction of new chunk types for tool management.
- Improved middleware architecture by integrating processing states and refining message handling, ensuring a more robust interaction with the API.
These changes enhance the overall maintainability and functionality of the API client and middleware, providing a more efficient framework for AI interactions.
* fix(McpToolChunkMiddleware): remove redundant logging in recursion state update
* refactor(McpToolChunkMiddleware): update tool call handling and type definitions
- Replaced ChatCompletionMessageToolCall with SdkToolCall for improved type consistency.
- Updated return types of executeToolCalls and executeToolUses functions to SdkMessage[], enhancing clarity in message handling.
- Removed unused import to streamline the code.
These changes enhance the maintainability and type safety of the middleware, ensuring better integration with the SDK.
* refactor(middleware): enhance middleware structure and type handling
- Updated middleware components to utilize new SDK types, improving type safety and clarity across the board.
- Refactored various middleware to streamline processing logic, including enhanced handling of SDK messages and tool calls.
- Improved logging and error handling for better debugging and maintainability.
- Consolidated middleware functions to reduce redundancy and improve overall architecture.
These changes enhance the robustness and maintainability of the middleware framework, ensuring a more efficient interaction with the API.
* refactor(middleware): unify type imports and enhance middleware structure
- Updated middleware components to import types from a unified 'types' file, improving consistency and clarity across the codebase.
- Removed the deprecated 'type.ts' file to streamline the middleware structure.
- Enhanced middleware registration and export mechanisms for better accessibility and maintainability.
These changes contribute to a more organized and efficient middleware framework, facilitating easier future development and integration.
* refactor(AiProvider): enhance API client and middleware integration
- Updated AiProvider components to support new SDK types, improving type safety and clarity.
- Refactored middleware to streamline processing logic, including enhanced handling of tool calls and responses.
- Introduced new middleware for tool use extraction and raw stream listening, improving overall functionality.
- Improved logging and error handling for better debugging and maintainability.
These changes enhance the robustness and maintainability of the API client and middleware, ensuring a more efficient interaction with the API.
* feat(middleware): add new middleware components for raw stream listening and tool use extraction
- Introduced RawStreamListenerMiddleware and ToolUseExtractionMiddleware to enhance middleware capabilities.
- Updated MiddlewareRegistry to include new middleware entries, improving overall functionality and extensibility.
These changes expand the middleware framework, facilitating better handling of streaming and tool usage scenarios.
* refactor(AiProvider): integrate new API client and middleware architecture
- Replaced BaseProvider with ApiClientFactory to enhance API client instantiation.
- Updated completions method to utilize new middleware architecture for improved processing.
- Added TODOs for refactoring remaining methods to align with the new API client structure.
- Removed deprecated middleware wrapping logic from ApiClientFactory for cleaner implementation.
These changes improve the overall structure and maintainability of the AiProvider, facilitating better integration with the new middleware system.
* refactor(middleware): update middleware architecture and documentation
- Revised middleware naming conventions and introduced a centralized MiddlewareRegistry for better management and accessibility.
- Enhanced MiddlewareBuilder to support named middleware and streamline the construction of middleware chains.
- Updated documentation to reflect changes in middleware usage and structure, improving clarity for future development.
These changes improve the organization and usability of the middleware framework, facilitating easier integration and maintenance.
* refactor(AiProvider): enhance completions middleware logic and API client handling
- Updated the completions method to conditionally remove middleware based on parameters, improving flexibility in processing.
- Refactored the response chunk transformer in OpenAIApiClient and AnthropicAPIClient to utilize a more streamlined approach with TransformStream.
- Simplified middleware context handling by removing unnecessary custom state management.
- Improved logging and error handling across middleware components for better debugging and maintainability.
These changes enhance the efficiency and clarity of the AiProvider's middleware integration, ensuring a more adaptable and robust processing framework.
* refactor(AiProvider, middleware): clean up logging and improve method naming
- Removed unnecessary logging of parameters in AiProvider to streamline the code.
- Updated method name assignment in middleware to enhance clarity and consistency.
These changes contribute to a cleaner codebase and improve the readability of the middleware and provider components.
* feat(middleware): enhance middleware types and add RawStreamListenerMiddleware
- Introduced RawStreamListenerMiddleware to the MiddlewareName enum for improved middleware capabilities.
- Updated type definitions across middleware components to enhance type safety and clarity, including the addition of new SDK types.
- Refactored context and middleware API interfaces to support more specific type parameters, improving overall maintainability.
These changes expand the middleware framework, facilitating better handling of streaming scenarios and enhancing type safety across the codebase.
* refactor(messageThunk): convert callback functions to async and handle errors during database updates
This commit updates several callback functions in the messageThunk to be asynchronous, ensuring that block transitions are awaited properly. Additionally, error handling is added for the database update function to log any failures when saving blocks. This improves the reliability and responsiveness of the message processing flow.
* refactor: enhance message block handling in messageThunk
This commit refactors the message processing logic in messageThunk to improve the management of message blocks. Key changes include the introduction of dedicated IDs for different block types (main text, thinking, tool, and image) to streamline updates and transitions. The handling of placeholder blocks has been improved, ensuring that they are correctly converted to their respective types during processing. Additionally, error handling has been enhanced for better reliability in database updates.
* feat(AiProvider): add default timeout configuration and enhance API client aborthandler
- Introduced a default timeout constant to the configuration for improved API client timeout management.
- Updated BaseApiClient and its derived classes to utilize the new timeout setting, ensuring consistent timeout behavior across different API clients.
- Enhanced middleware to pass the timeout value during API calls, improving error handling and responsiveness.
These changes improve the overall robustness and configurability of the API client interactions, facilitating better control over request timeouts.
* feat(GeminiProvider): implement Gemini API client and enhance file handling
- Introduced GeminiAPIClient to facilitate interactions with the Gemini API, replacing the previous GoogleGenAI integration.
- Refactored GeminiProvider to utilize the new API client, improving code organization and maintainability.
- Enhanced file handling capabilities, including support for PDF uploads and retrieval of file metadata.
- Updated message processing to accommodate new SDK types and improve content generation logic.
These changes significantly enhance the functionality and robustness of the GeminiProvider, enabling better integration with the Gemini API and improving overall user experience.
* refactor(AiProvider, middleware): streamline API client and middleware integration
- Removed deprecated methods and types from various API clients, enhancing code clarity and maintainability.
- Updated the CompletionsParams interface to support messages as a string or array, improving flexibility in message handling.
- Refactored middleware components to eliminate unnecessary state management and improve type safety.
- Enhanced the handling of streaming responses and added utility functions for better stream management.
These changes contribute to a more robust and efficient architecture for the AiProvider and its associated middleware, facilitating improved API interactions and user experience.
* refactor(middleware): translation 适配
- Deleted SdkCallMiddleware to streamline middleware architecture and improve maintainability.
- Commented out references to SdkCallModule in examples and registration files to prevent usage.
- Enhanced logging in AbortHandlerMiddleware for better debugging and tracking of middleware execution.
- Updated parameters in ResponseTransformMiddleware to improve flexibility in handling response settings.
These changes contribute to a cleaner and more efficient middleware framework, facilitating better integration and performance.
* refactor(ApiCheck): streamline API validation and error handling
- Updated the API check logic to simplify validation processes and improve error handling across various components.
- Refactored the `checkApi` function to throw errors directly instead of returning validation objects, enhancing clarity in error management.
- Improved the handling of API key checks in `checkModelWithMultipleKeys` to provide more informative error messages.
- Added a new method `getEmbeddingDimensions` in the `AiProvider` class to facilitate embedding dimension retrieval, enhancing model compatibility checks.
These changes contribute to a more robust and maintainable API validation framework, improving overall user experience and error reporting.
* refactor(HealthCheckService, ModelService): improve error handling and performance metrics
- Updated error handling in `checkModelWithMultipleKeys` to truncate error messages for better readability.
- Refactored `performModelCheck` to remove unnecessary error handling, focusing on performance metrics by returning only latency.
- Enhanced the `checkModel` function to ensure consistent return types, improving clarity in API interactions.
These changes contribute to a more efficient and user-friendly error reporting and performance tracking system.
* refactor(AiProvider, models): enhance model handling and API client integration
- Updated the `listModels` method in various API clients to improve model retrieval and ensure consistent return types.
- Refactored the `EditModelsPopup` component to handle model properties more robustly, including fallback options for `id`, `name`, and other attributes.
- Enhanced type definitions for models in the SDK to support new integrations and improve type safety.
These changes contribute to a more reliable and maintainable model management system within the AiProvider, enhancing overall user experience and API interactions.
* refactor(AiProvider, clients): implement image generation functionality
- Refactored the `generateImage` method in the `AiProvider` class to utilize the `apiClient` for image generation, replacing the previous placeholder implementation.
- Updated the `BaseApiClient` to include an abstract `generateImage` method, ensuring all derived clients implement this functionality.
- Implemented the `generateImage` method in `GeminiAPIClient` and `OpenAIAPIClient`, providing specific logic for image generation based on the respective SDKs.
- Added type definitions for `GenerateImageParams` across relevant files to enhance type safety and clarity in image generation parameters.
These changes enhance the image generation capabilities of the AiProvider, improving integration with various API clients and overall user experience.
* refactor(AiProvider, clients): restructure API client architecture and remove deprecated components
- Refactored the `ProviderFactory` and removed the `AihubmixProvider` to streamline the API client architecture.
- Updated the import paths for `isOpenAIProvider` to reflect the new structure.
- Introduced `AihubmixAPIClient` and `OpenAIResponseAPIClient` to enhance client handling based on model types.
- Improved the `AiProvider` class to utilize the new clients for better model-specific API interactions.
- Enhanced type definitions and error handling across various components to improve maintainability and clarity.
These changes contribute to a more efficient and organized API client structure, enhancing overall integration and user experience.
* fix: update system prompt handling in API clients to use await for asynchronous operations
- Modified the `AnthropicAPIClient`, `GeminiAPIClient`, `OpenAIAPIClient`, and `OpenAIResponseAPIClient` to ensure `buildSystemPrompt` is awaited, improving the handling of system prompts.
- Adjusted the `fetchMessagesSummary` function to utilize the last five user messages for better context in API calls and added a utility function to clean up topic names.
These changes enhance the reliability of prompt generation and improve the overall API interaction experience.
* refactor(middleware): remove examples.ts to streamline middleware documentation
- Deleted the `examples.ts` file containing various middleware usage examples to simplify the middleware structure and documentation.
- This change contributes to a cleaner codebase and focuses on essential middleware components, enhancing maintainability.
* refactor(AiProvider, middleware): enhance middleware handling and error management
- Updated the `CompletionsParams` interface to include a new `callType` property for better middleware decision-making based on the context of the API call.
- Introduced `ErrorHandlerMiddleware` to standardize error handling across middleware, allowing errors to be captured and processed as `ErrorChunk` objects.
- Modified the `AbortHandlerMiddleware` to conditionally remove itself based on the `callType`, improving middleware efficiency.
- Cleaned up logging in `AbortHandlerMiddleware` to reduce console output and enhance performance.
- Updated middleware registration to include the new `ErrorHandlerMiddleware`, ensuring comprehensive error management in the middleware pipeline.
These changes contribute to a more robust and maintainable middleware architecture, improving error handling and overall API interaction efficiency.
* feat: implement token estimation for message handling
- Added an abstract method `estimateMessageTokens` to the `BaseApiClient` class for estimating token usage based on message content.
- Implemented the `estimateMessageTokens` method in `AnthropicAPIClient`, `GeminiAPIClient`, `OpenAIAPIClient`, and `OpenAIResponseAPIClient` to calculate token consumption for various message types.
- Enhanced middleware to accumulate token usage for new messages, improving tracking of API call costs.
These changes improve the efficiency of message processing and provide better insights into token usage across different API clients.
* feat: add support for image generation and model handling
- Introduced `SUPPORTED_DISABLE_GENERATION_MODELS` to manage models that disable image generation.
- Updated `isSupportedDisableGenerationModel` function to check model compatibility.
- Enhanced `Inputbar` logic to conditionally enable image generation based on model support.
- Modified API clients to handle image generation calls and responses, including new chunk types for image data.
- Updated middleware and service layers to incorporate image generation parameters and improve overall processing.
These changes enhance the application's capabilities for image generation and improve the handling of various model types.
* feat: enhance GeminiAPIClient for image generation support
- Added `getGenerateImageParameter` method to configure image generation parameters.
- Updated request handling in `GeminiAPIClient` to include image generation options.
- Enhanced response processing to handle image data and enqueue it correctly.
These changes improve the GeminiAPIClient's capabilities for generating and processing images, aligning with recent enhancements in image generation support.
* feat: enhance image generation handling in OpenAIResponseAPIClient and middleware
- Updated OpenAIResponseAPIClient to improve user message processing for image generation.
- Added handling for image creation events in TransformCoreToSdkParamsMiddleware.
- Adjusted ApiService to streamline image generation event handling.
- Modified messageThunk to reflect changes in image block status during processing.
These enhancements improve the integration and responsiveness of image generation features across the application.
* refactor: remove unused AI provider classes
- Deleted `AihubmixProvider`, `AnthropicProvider`, `BaseProvider`, `GeminiProvider`, and `OpenAIProvider` as they are no longer utilized in the codebase.
- This cleanup reduces code complexity and improves maintainability by removing obsolete components related to AI provider functionality.
* chore: remove obsolete test files for middleware
- Deleted test files for `AbortHandlerMiddleware`, `LoggingMiddleware`, `TextChunkMiddleware`, `ThinkChunkMiddleware`, and `WebSearchMiddleware` as they are no longer needed.
- This cleanup helps streamline the codebase and reduces maintenance overhead by removing outdated tests.
* chore: remove Suggestions component and related functionality
- Deleted the `Suggestions` component from the home page as it is no longer needed.
- Removed associated imports and functions related to suggestion fetching, streamlining the codebase.
- This cleanup helps improve maintainability by eliminating unused components.
* feat: enhance OpenAIAPIClient and StreamProcessingService for tool call handling
- Updated OpenAIAPIClient to conditionally include tool calls in the assistant message, improving message processing logic.
- Enhanced tool call handling in the response transformer to correctly manage and enqueue tool call data.
- Added a new callback for LLM response completion in StreamProcessingService, allowing better integration of response handling.
These changes improve the functionality and responsiveness of the OpenAI API client and stream processing capabilities.
* fix: copilot error
* fix: improve chunk handling in TextChunkMiddleware and ThinkChunkMiddleware
- Updated TextChunkMiddleware to enqueue LLM_RESPONSE_COMPLETE chunks based on accumulated text content.
- Refactored ThinkChunkMiddleware to generate THINKING_COMPLETE chunks when receiving non-THINKING_DELTA chunks, ensuring proper handling of accumulated thinking content.
- These changes enhance the middleware's responsiveness and accuracy in processing text and thinking chunks.
* chore: update dependencies and improve styling
- Updated `selection-hook` dependency to version 0.9.23 in `package.json` and `yarn.lock`.
- Removed unused styles from `container.scss` and adjusted padding in `index.scss`.
- Enhanced message rendering and layout in various components, including `Message`, `MessageHeader`, and `MessageMenubar`.
- Added tooltip support for message divider settings in `SettingsTab`.
- Improved handling of citation display in `CitationsList` and `CitationBlock`.
These changes streamline the codebase and enhance the user interface for better usability.
* feat: implement image generation middleware and enhance model handling
- Added `ImageGenerationMiddleware` to handle dedicated image generation models, integrating image processing and OpenAI's image generation API.
- Updated `AiProvider` to utilize the new middleware for dedicated image models, ensuring proper middleware chaining.
- Introduced constants for dedicated image models in `models.ts` to streamline model identification.
- Refactored error handling in `ErrorHandlerMiddleware` to use a utility function for better error management.
- Cleaned up imports and removed unused code in various files for improved maintainability.
* fix: update dedicated image models identification logic
- Modified the `DEDICATED_IMAGE_MODELS` array to include 'grok-2-image' for improved model handling.
- Enhanced the `isDedicatedImageGenerationModel` function to use a more robust check for model identification, ensuring better accuracy in middleware processing.
* refactor: remove OpenAIResponseProvider class
- Deleted the `OpenAIResponseProvider` class from the `AiProvider` module, streamlining the codebase by eliminating unused code.
- This change enhances maintainability and reduces complexity in the provider architecture.
* fix: usermessage
* refactor: simplify AbortHandlerMiddleware for improved abort handling
- Removed direct dependency on ApiClient for creating AbortController, enhancing modularity.
- Introduced utility functions to manage abort controllers, streamlining the middleware's responsibilities.
- Delegated abort signal handling to downstream middlewares, allowing for cleaner separation of concerns.
* refactor(aiCore): Consolidate AI provider and middleware architecture
This commit refactors the AI-related modules by unifying the `clients` and `middleware` directories under a single `aiCore` directory. This change simplifies the project structure, improves modularity, and makes the architecture more cohesive.
Key changes:
- Relocated provider-specific clients and middleware into the `aiCore` directory, removing the previous `providers/AiProvider` structure.
- Updated the architectural documentation (`AI_CORE_DESIGN.md`) to accurately reflect the new, streamlined directory layout and execution flow.
- The main `AiProvider` class is now the primary export of `aiCore/index.ts`, serving as the central access point for AI functionalities.
* refactor: update imports and enhance middleware functionality
- Adjusted import statements in `AnthropicAPIClient` and `GeminiAPIClient` for better organization.
- Improved `AbortHandlerMiddleware` to handle abort signals more effectively, including the conversion of streams to handle abort scenarios.
- Enhanced `ErrorHandlerMiddleware` to differentiate between abort errors and other types, ensuring proper error handling.
- Cleaned up commented-out code in `FinalChunkConsumerMiddleware` for better readability and maintainability.
* refactor: streamline middleware logging and improve error handling
- Removed excessive debug logging from various middleware components, including `AbortHandlerMiddleware`, `FinalChunkConsumerMiddleware`, and `McpToolChunkMiddleware`, to enhance readability and performance.
- Updated logging levels to use warnings for potential issues in `ResponseTransformMiddleware`, `TextChunkMiddleware`, and `ThinkChunkMiddleware`, ensuring better visibility of important messages.
- Cleaned up commented-out code and unnecessary debug statements across multiple middleware files for improved maintainability.
---------
Co-authored-by: suyao <sy20010504@gmail.com>
Co-authored-by: eeee0717 <chentao020717Work@outlook.com>
Co-authored-by: lizhixuan <zhixuan.li@banosuperapp.com>
2025-06-12 16:01:19 +08:00
752 changed files with 72625 additions and 13701 deletions
index 433e2efc9cef156ff5444f0c4520362ed2ef9ea7..a89c7a9b0b608fef67902c49106a43ebd0fa8b61 100644
index 433e2efc9cef156ff5444f0c4520362ed2ef9ea7..0167441bf928a92f59b5dbe70b2317a74dda74c9 100644
--- a/scheme.json
+++ b/scheme.json
@@ -1975,6 +1975,13 @@
@@ -1825,6 +1825,20 @@
"string"
]
},
+ "excludeReBuildModules": {
+ "anyOf": [
+ {
+ "items": {
+ "type": "string"
+ },
+ "type": "array"
+ },
+ {
+ "type": "null"
+ }
+ ],
+ "description": "The modules to exclude from the rebuild."
+ },
"executableArgs": {
"anyOf": [
{
@@ -1975,6 +1989,13 @@
],
"description": "The mime types in addition to specified in the file associations. Use it if you don't want to register a new mime type, but reuse existing."
},
@@ -83,7 +116,7 @@ index 433e2efc9cef156ff5444f0c4520362ed2ef9ea7..a89c7a9b0b608fef67902c49106a43eb
"packageCategory": {
"description": "backward compatibility + to allow specify fpm-only category for all possible fpm targets in one place",
"type": [
@@ -2327,6 +2334,13 @@
@@ -2327,6 +2348,13 @@
"MacConfiguration": {
"additionalProperties": false,
"properties": {
@@ -97,7 +130,28 @@ index 433e2efc9cef156ff5444f0c4520362ed2ef9ea7..a89c7a9b0b608fef67902c49106a43eb
"additionalArguments": {
"anyOf": [
{
@@ -2737,7 +2751,7 @@
@@ -2527,6 +2555,20 @@
"string"
]
},
+ "excludeReBuildModules": {
+ "anyOf": [
+ {
+ "items": {
+ "type": "string"
+ },
+ "type": "array"
+ },
+ {
+ "type": "null"
+ }
+ ],
+ "description": "The modules to exclude from the rebuild."
+ },
"executableName": {
"description": "The executable name. Defaults to `productName`.",
"type": [
@@ -2737,7 +2779,7 @@
"type": "boolean"
},
"minimumSystemVersion": {
@@ -106,7 +160,7 @@ index 433e2efc9cef156ff5444f0c4520362ed2ef9ea7..a89c7a9b0b608fef67902c49106a43eb
"type": [
"null",
"string"
@@ -2959,6 +2973,13 @@
@@ -2959,6 +3001,13 @@
"MasConfiguration": {
"additionalProperties": false,
"properties": {
@@ -120,7 +174,28 @@ index 433e2efc9cef156ff5444f0c4520362ed2ef9ea7..a89c7a9b0b608fef67902c49106a43eb
"additionalArguments": {
"anyOf": [
{
@@ -3369,7 +3390,7 @@
@@ -3159,6 +3208,20 @@
"string"
]
},
+ "excludeReBuildModules": {
+ "anyOf": [
+ {
+ "items": {
+ "type": "string"
+ },
+ "type": "array"
+ },
+ {
+ "type": "null"
+ }
+ ],
+ "description": "The modules to exclude from the rebuild."
+ },
"executableName": {
"description": "The executable name. Defaults to `productName`.",
"type": [
@@ -3369,7 +3432,7 @@
"type": "boolean"
},
"minimumSystemVersion": {
@@ -129,7 +204,28 @@ index 433e2efc9cef156ff5444f0c4520362ed2ef9ea7..a89c7a9b0b608fef67902c49106a43eb
"type": [
"null",
"string"
@@ -6507,6 +6528,13 @@
@@ -6381,6 +6444,20 @@
"string"
]
},
+ "excludeReBuildModules": {
+ "anyOf": [
+ {
+ "items": {
+ "type": "string"
+ },
+ "type": "array"
+ },
+ {
+ "type": "null"
+ }
+ ],
+ "description": "The modules to exclude from the rebuild."
+ },
"executableName": {
"description": "The executable name. Defaults to `productName`.",
"type": [
@@ -6507,6 +6584,13 @@
"string"
]
},
@@ -143,7 +239,28 @@ index 433e2efc9cef156ff5444f0c4520362ed2ef9ea7..a89c7a9b0b608fef67902c49106a43eb
"protocols": {
"anyOf": [
{
@@ -7376,6 +7404,13 @@
@@ -7153,6 +7237,20 @@
"string"
]
},
+ "excludeReBuildModules": {
+ "anyOf": [
+ {
+ "items": {
+ "type": "string"
+ },
+ "type": "array"
+ },
+ {
+ "type": "null"
+ }
+ ],
+ "description": "The modules to exclude from the rebuild."
+ },
"executableName": {
"description": "The executable name. Defaults to `productName`.",
"type": [
@@ -7376,6 +7474,13 @@
],
"description": "MAS (Mac Application Store) development options (`mas-dev` target)."
Maintainers are here to help you implement your use case within a reasonable timeframe. They will do their best to review your code and provide constructive feedback promptly. However, if you get stuck during the review process or feel your Pull Request is not receiving the attention it deserves, please contact us via comments in the Issue or through the [Community](README.md#-community).
### Participating in the Test Plan
The Test Plan aims to provide users with a more stable application experience and faster iteration speed. For details, please refer to the [Test Plan](docs/testplan-en.md).
### Other Suggestions
- **Contact Developers**: Before submitting a PR, you can contact the developers first to discuss or get help.
@@ -163,10 +183,82 @@ Refer to the [Branching Strategy](docs/branching-strategy-en.md) for contributio
3.**Submit Changes**: Commit and push your changes.
4.**Open a Pull Request**: Describe your changes and reasons.
For more detailed guidelines, please refer to our [Contributing Guide](./CONTRIBUTING.md).
For more detailed guidelines, please refer to our [Contributing Guide](CONTRIBUTING.md).
Thank you for your support and contributions!
# 🔧 Developer Co-creation Program
We are launching the Cherry Studio Developer Co-creation Program to foster a healthy and positive-feedback loop within the open-source ecosystem. We believe that great software is built collaboratively, and every merged pull request breathes new life into the project.
We sincerely invite you to join our ranks of contributors and shape the future of Cherry Studio with us.
## Contributor Rewards Program
To give back to our core contributors and create a virtuous cycle, we have established the following long-term incentive plan.
**The inaugural tracking period for this program will be Q3 2025 (July, August, September). Rewards for this cycle will be distributed on October 1st.**
Within any tracking period (e.g., July 1st to September 30th for the first cycle), any developer who contributes more than **30 meaningful commits** to any of Cherry Studio's open-source projects on GitHub is eligible for the following benefits:
- **Cursor Subscription Sponsorship**: Receive a **$70 USD** credit or reimbursement for your [Cursor](https://cursor.sh/) subscription, making AI your most efficient coding partner.
- **Unlimited Model Access**: Get **unlimited** API calls for the **DeepSeek** and **Qwen** models.
- **Cutting-Edge Tech Access**: Enjoy occasional perks, including API access to models like **Claude**, **Gemini**, and **OpenAI**, keeping you at the forefront of technology.
## Growing Together & Future Plans
A vibrant community is the driving force behind any sustainable open-source project. As Cherry Studio grows, so will our rewards program. We are committed to continuously aligning our benefits with the best-in-class tools and resources in the industry. This ensures our core contributors receive meaningful support, creating a positive cycle where developers, the community, and the project grow together.
**Moving forward, the project will also embrace an increasingly open stance to give back to the entire open-source community.**
## How to Get Started?
We look forward to your first Pull Request!
You can start by exploring our repositories, picking up a `good first issue`, or proposing your own enhancements. Every commit is a testament to the spirit of open source.
Thank you for your interest and contributions.
Let's build together.
# 🏢 Enterprise Edition
Building on the Community Edition, we are proud to introduce **Cherry Studio Enterprise Edition**—a privately deployable AI productivity and management platform designed for modern teams and enterprises.
The Enterprise Edition addresses core challenges in team collaboration by centralizing the management of AI resources, knowledge, and data. It empowers organizations to enhance efficiency, foster innovation, and ensure compliance, all while maintaining 100% control over their data in a secure environment.
## Core Advantages
- **Unified Model Management**: Centrally integrate and manage various cloud-based LLMs (e.g., OpenAI, Anthropic, Google Gemini) and locally deployed private models. Employees can use them out-of-the-box without individual configuration.
- **Enterprise-Grade Knowledge Base**: Build, manage, and share team-wide knowledge bases. Ensure knowledge is retained and consistent, enabling team members to interact with AI based on unified and accurate information.
- **Fine-Grained Access Control**: Easily manage employee accounts and assign role-based permissions for different models, knowledge bases, and features through a unified admin backend.
- **Fully Private Deployment**: Deploy the entire backend service on your on-premises servers or private cloud, ensuring your data remains 100% private and under your control to meet the strictest security and compliance standards.
- **Reliable Backend Services**: Provides stable API services, enterprise-grade data backup and recovery mechanisms to ensure business continuity.
## ✨ Online Demo
> 🚧 **Public Beta Notice**
>
> The Enterprise Edition is currently in its early public beta stage, and we are actively iterating and optimizing its features. We are aware that it may not be perfectly stable yet. If you encounter any issues or have valuable suggestions during your trial, we would be very grateful if you could contact us via email to provide feedback.
**🔗 [Cherry Studio Enterprise](https://www.cherry-ai.com/enterprise)**
## Version Comparison
| Feature | Community Edition | Enterprise Edition |
We believe the Enterprise Edition will become your team's AI productivity engine. If you are interested in Cherry Studio Enterprise Edition and would like to learn more, request a quote, or schedule a demo, please contact us.
- **For Business Inquiries & Purchasing**:
**📧 [bd@cherry-ai.com](mailto:bd@cherry-ai.com)**
# 🔗 Related Projects
- [one-api](https://github.com/songquanpeng/one-api):LLM API management and distribution system, supporting mainstream models like OpenAI, Azure, and Anthropic. Features unified API interface, suitable for key management and secondary distribution.
@@ -180,34 +272,45 @@ Thank you for your support and contributions!
[](https://star-history.com/#CherryHQ/cherry-studio&Timeline)
[](https://star-history.com/#CherryHQ/cherry-studio&Timeline)
[](https://star-history.com/#CherryHQ/cherry-studio&Timeline)
To provide users with a more stable application experience and faster iteration speed, Cherry Studio has launched the "Test Plan".
## User Guide
The Test Plan is divided into the RC channel and the Beta channel, with the following differences:
- **RC (Release Candidate)**: The features are stable, with fewer bugs, and it is close to the official release.
- **Beta**: Features may change at any time, and there may be more bugs, but users can experience future features earlier.
Users can enable the "Test Plan" and select the version channel in the software's `Settings` > `About`. Please note that the versions in the "Test Plan" cannot guarantee data consistency, so be sure to back up your data before using them.
Users are welcome to submit issues or provide feedback through other channels for any bugs encountered during testing. Your feedback is very important to us.
## Developer Guide
### Participating in the Test Plan
Developers should submit `PRs` according to the [Contributor Guide](../CONTRIBUTING.md) (and ensure the target branch is `main`). The repository maintainers will evaluate whether the `PR` should be included in the Test Plan based on factors such as the impact of the feature on the application, its importance, and whether broader testing is needed.
If the `PR` is added to the Test Plan, the repository maintainers will:
- Notify the `PR` submitter.
- Set the PR to `draft` status (to avoid accidental merging into `main` before testing is complete).
- Set the `milestone` to the specific Test Plan version.
- Modify the `PR` title.
During participation in the Test Plan, `PR` submitters should:
- Keep the `PR` branch synchronized with the latest `main` (i.e., the `PR` branch should always be based on the latest `main` code).
- Ensure the `PR` branch is conflict-free.
- Actively respond to comments & reviews and fix bugs.
- Enable maintainers to modify the `PR` branch to allow for bug fixes at any time.
Inclusion in the Test Plan does not guarantee the final merging of the `PR`. It may be shelved due to immature features or poor testing feedback.
### Test Plan Lead
A maintainer will be assigned as the lead for a specific version (e.g., `1.5.0-rc`). The responsibilities of the Test Plan lead include:
- Determining whether a `PR` meets the Test Plan requirements and deciding whether it should be included in the current Test Plan.
- Modifying the status of `PRs` added to the Test Plan and communicating relevant matters with the `PR` submitter.
- Before the Test Plan release, merging the branches of `PRs` added to the Test Plan (using squash merge) into the corresponding version branch of `testplan` and resolving conflicts.
- Ensuring the `testplan` branch is synchronized with the latest `main`.
- Overseeing the Test Plan release.
## In-Depth Understanding
### About `PRs`
A `PR` is a collection of a specific branch (and commits), comments, reviews, and other information, and it is the **smallest management unit** of the Test Plan.
Compared to submitting all features to a single branch, the Test Plan manages features through `PRs`, which offers greater flexibility and efficiency:
- Features can be added or removed between different versions of the Test Plan without cumbersome `revert` operations.
- Clear feature boundaries and responsibilities are established. Bug fixes are completed within their respective `PRs`, isolating cross-impact and better tracking progress.
- The `PR` submitter is responsible for resolving conflicts with the latest `main`. The Test Plan lead is responsible for resolving conflicts between `PR` branches. However, since features added to the Test Plan are relatively independent (in other words, if a feature has broad implications, it should be independently included in the Test Plan), conflicts are generally few or simple.
### The `testplan` Branch
The `testplan` branch is a **temporary** branch used for Test Plan releases.
Note:
- **Do not develop based on this branch**. It may change or even be deleted at any time, and there is no guarantee of commit completeness or order.
- **Do not submit `commits` or `PRs` to this branch**, as they will not be retained.
- The `testplan` branch is always based on the latest `main` branch (not on a released version), with features added on top.
#### RC Branch
Branch name: `testplan/rc/x.y.z`
Used for RC releases, where `x.y.z` is the target version number. Note that whether it is rc.1 or rc.5, as long as the major version number is `x.y.z`, it is completed in this branch.
Generally, the version number for releases from this branch is named `x.y.z-rc.n`.
#### Beta Branch
Branch name: `testplan/beta/x.y.z`
Used for Beta releases, where `x.y.z` is the target version number. Note that whether it is beta.1 or beta.5, as long as the major version number is `x.y.z`, it is completed in this branch.
Generally, the version number for releases from this branch is named `x.y.z-beta.n`.
### Version Rules
The application version number for the Test Plan is: `x.y.z-CHA.n`, where:
-`x.y.z` is the conventional version number, referred to here as the **target version number**.
-`CHA` is the channel code (Channel), currently divided into `rc` and `beta`.
-`n` is the release number, starting from `1`.
Examples of complete version numbers: `1.5.0-rc.3`, `1.5.1-beta.1`, `1.6.0-beta.6`.
The **target version number** of the Test Plan points to the official version number where these features are expected to be added. For example:
-`1.5.0-rc.3` means this is a preview of the `1.5.0` official release (the current latest official release is `1.4.9`, and `1.5.0` has not yet been officially released).
-`1.5.1-beta.1` means this is a beta version of the `1.5.1` official release (the current latest official release is `1.5.0`, and `1.5.1` has not yet been officially released).
The Yi Large model was designed by 01.AI with the following usecases in mind: knowledge search, data classification, human-like chat bots, and customer service.
It stands out for its multilingual proficiency, particularly in Spanish, Chinese, Japanese, German, and French.
Check out the [launch announcement](https://01-ai.github.io/blog/01.ai-yi-large-llm-launch) to learn more.
Starcannon 12B v2 is a creative roleplay and story writing model, based on Mistral Nemo, using [nothingiisreal/mn-celeste-12b](/nothingiisreal/mn-celeste-12b) as a base, with [intervitens/mini-magnum-12b-v1.1](https://huggingface.co/intervitens/mini-magnum-12b-v1.1) merged in using the [TIES](https://arxiv.org/abs/2306.01708) method.
Although more similar to Magnum overall, the model remains very creative, with a pleasant writing style. It is recommended for people wanting more variety than Magnum, and yet more verbose prose than Celeste.
AI21 Jamba Large 1.6 is a high-performance hybrid foundation model combining State Space Models (Mamba) with Transformer attention mechanisms. Developed by AI21, it excels in extremely long-context handling (256K tokens), demonstrates superior inference efficiency (up to 2.5x faster than comparable models), and supports structured JSON output and tool-use capabilities. It has 94 billion active parameters (398 billion total), optimized quantization support (ExpertsInt8), and multilingual proficiency in languages such as English, Spanish, French, Portuguese, Italian, Dutch, German, Arabic, and Hebrew.
Usage of this model is subject to the [Jamba Open Model License](https://www.ai21.com/licenses/jamba-open-model-license).
AI21 Jamba Mini 1.6 is a hybrid foundation model combining State Space Models (Mamba) with Transformer attention mechanisms. With 12 billion active parameters (52 billion total), this model excels in extremely long-context tasks (up to 256K tokens) and achieves superior inference efficiency, outperforming comparable open models on tasks such as retrieval-augmented generation (RAG) and grounded question answering. Jamba Mini 1.6 supports multilingual tasks across English, Spanish, French, Portuguese, Italian, Dutch, German, Arabic, and Hebrew, along with structured JSON output and tool-use capabilities.
Usage of this model is subject to the [Jamba Open Model License](https://www.ai21.com/licenses/jamba-open-model-license).
description:Aion-1.0-Mini 32B parameter model is a distilled version of the DeepSeek-R1 model, designed for strong performance in reasoning domains such as mathematics, coding, and logic. It is a modified variant of a FuseAI model that outperforms R1-Distill-Qwen-32B and R1-Distill-Llama-70B, with benchmark results available on its [Hugging Face page](https://huggingface.co/FuseAI/FuseO1-DeepSeekR1-QwQ-SkyT1-32B-Preview), independently replicated for verification.
description:Aion-1.0 is a multi-model system designed for high performance across various tasks, including reasoning and coding. It is built on DeepSeek-R1, augmented with additional models and techniques such as Tree of Thoughts (ToT) and Mixture of Experts (MoE). It is Aion Lab's most powerful reasoning model.
description:Aion-RP-Llama-3.1-8B ranks the highest in the character evaluation portion of the RPBench-Auto benchmark, a roleplaying-specific variant of Arena-Hard-Auto, where LLMs evaluate each other’s responses. It is a fine-tuned base model rather than an instruct model, designed to produce more natural and varied writing.
OpenHands LM v0.1 is a 32B open-source coding model fine-tuned from Qwen2.5-Coder-32B-Instruct using reinforcement learning techniques outlined in SWE-Gym. It is optimized for autonomous software development agents and achieves strong performance on SWE-Bench Verified, with a 37.2% resolve rate. The model supports a 128K token context window, making it well-suited for long-horizon code reasoning and large codebase tasks.
OpenHands LM is designed for local deployment and runs on consumer-grade GPUs such as a single 3090. It enables fully offline agent workflows without dependency on proprietary APIs. This release is intended as a research preview, and future updates aim to improve generalizability, reduce repetition, and offer smaller variants.
A large LLM created by combining two fine-tuned Llama 70B models into one 120B model. Combines Xwin and Euryale.
Credits to
- [@chargoddard](https://huggingface.co/chargoddard) for developing the framework used to merge the model - [mergekit](https://github.com/cg123/mergekit).
- [@Undi95](https://huggingface.co/Undi95) for helping with the merge ratios.
From the maker of [Goliath](https://openrouter.ai/models/alpindale/goliath-120b), Magnum 72B is the first in a new family of models designed to achieve the prose quality of the Claude 3 models, notably Opus & Sonnet.
The model is based on [Qwen2 72B](https://openrouter.ai/models/qwen/qwen-2-72b-instruct) and trained with 55 million tokens of highly curated roleplay (RP) data.
Amazon Nova Lite 1.0 is a very low-cost multimodal model from Amazon that focused on fast processing of image, video, and text inputs to generate text output. Amazon Nova Lite can handle real-time customer interactions, document analysis, and visual question-answering tasks with high accuracy.
With an input context of 300K tokens, it can analyze multiple images or up to 30 minutes of video in a single input.
description:Amazon Nova Micro 1.0 is a text-only model that delivers the lowest latency responses in the Amazon Nova family of models at a very low cost. With a context length of 128K tokens and optimized for speed and cost, Amazon Nova Micro excels at tasks such as text summarization, translation, content classification, interactive chat, and brainstorming. It has simple mathematical reasoning and coding abilities.
Amazon Nova Pro 1.0 is a capable multimodal model from Amazon focused on providing a combination of accuracy, speed, and cost for a wide range of tasks. As of December 2024, it achieves state-of-the-art performance on key benchmarks including visual question answering (TextVQA) and video understanding (VATEX).
Amazon Nova Pro demonstrates strong capabilities in processing both visual and textual information and at analyzing financial documents.
**NOTE**: Video input is not supported at this time.
From the maker of [Goliath](https://openrouter.ai/models/alpindale/goliath-120b), Magnum 72B is the seventh in a family of models designed to achieve the prose quality of the Claude 3 models, notably Opus & Sonnet.
The model is based on [Qwen2 72B](https://openrouter.ai/models/qwen/qwen-2-72b-instruct) and trained with 55 million tokens of highly curated roleplay (RP) data.
This is a series of models designed to replicate the prose quality of the Claude 3 models, specifically Sonnet(https://openrouter.ai/anthropic/claude-3.5-sonnet) and Opus(https://openrouter.ai/anthropic/claude-3-opus).
The model is fine-tuned on top of [Qwen2.5 72B](https://openrouter.ai/qwen/qwen-2.5-72b-instruct).
description: 'Claude 2 delivers advancements in key capabilities for enterprises—including an industry-leading 200K token context window, significant reductions in rates of model hallucination, system prompts and a new beta feature:tool use.'
description: 'Claude 2 delivers advancements in key capabilities for enterprises—including an industry-leading 200K token context window, significant reductions in rates of model hallucination, system prompts and a new beta feature:tool use.'
description: 'Claude 2 delivers advancements in key capabilities for enterprises—including an industry-leading 200K token context window, significant reductions in rates of model hallucination, system prompts and a new beta feature:tool use.'
description: 'Claude 2 delivers advancements in key capabilities for enterprises—including an industry-leading 200K token context window, significant reductions in rates of model hallucination, system prompts and a new beta feature:tool use.'
Claude 3 Sonnet is an ideal balance of intelligence and speed for enterprise workloads. Maximum utility at a lower price, dependable, balanced for scaled deployments.
See the launch announcement and benchmark results [here](https://www.anthropic.com/news/claude-3-family)
Claude 3 Sonnet is an ideal balance of intelligence and speed for enterprise workloads. Maximum utility at a lower price, dependable, balanced for scaled deployments.
See the launch announcement and benchmark results [here](https://www.anthropic.com/news/claude-3-family)
Claude 3.5 Haiku features enhancements across all skill sets including coding, tool use, and reasoning. As the fastest model in the Anthropic lineup, it offers rapid response times suitable for applications that require high interactivity and low latency, such as user-facing chatbots and on-the-fly code completions. It also excels in specialized tasks like data extraction and real-time content moderation, making it a versatile tool for a broad range of industries.
It does not support image inputs.
See the launch announcement and benchmark results [here](https://www.anthropic.com/news/3-5-models-and-computer-use)
Claude 3.5 Haiku features enhancements across all skill sets including coding, tool use, and reasoning. As the fastest model in the Anthropic lineup, it offers rapid response times suitable for applications that require high interactivity and low latency, such as user-facing chatbots and on-the-fly code completions. It also excels in specialized tasks like data extraction and real-time content moderation, making it a versatile tool for a broad range of industries.
It does not support image inputs.
See the launch announcement and benchmark results [here](https://www.anthropic.com/news/3-5-models-and-computer-use)
Claude 3.5 Haiku features offers enhanced capabilities in speed, coding accuracy, and tool use. Engineered to excel in real-time applications, it delivers quick response times that are essential for dynamic tasks such as chat interactions and immediate coding suggestions.
This makes it highly suitable for environments that demand both speed and precision, such as software development, customer service bots, and data management systems.
This model is currently pointing to [Claude 3.5 Haiku (2024-10-22)](/anthropic/claude-3-5-haiku-20241022).
Claude 3.5 Haiku features offers enhanced capabilities in speed, coding accuracy, and tool use. Engineered to excel in real-time applications, it delivers quick response times that are essential for dynamic tasks such as chat interactions and immediate coding suggestions.
This makes it highly suitable for environments that demand both speed and precision, such as software development, customer service bots, and data management systems.
This model is currently pointing to [Claude 3.5 Haiku (2024-10-22)](/anthropic/claude-3-5-haiku-20241022).
Claude 3.5 Sonnet delivers better-than-Opus capabilities, faster-than-Sonnet speeds, at the same Sonnet prices. Sonnet is particularly good at:
- Coding: Autonomously writes, edits, and runs code with reasoning and troubleshooting
- Data science: Augments human data science expertise; navigates unstructured data while using multiple tools for insights
- Visual processing: excelling at interpreting charts, graphs, and images, accurately transcribing text to derive insights beyond just the text alone
- Agentic tasks: exceptional tool use, making it great at agentic tasks (i.e. complex, multi-step problem solving tasks that require engaging with other systems)
For the latest version (2024-10-23), check out [Claude 3.5 Sonnet](/anthropic/claude-3.5-sonnet).
Claude 3.5 Sonnet delivers better-than-Opus capabilities, faster-than-Sonnet speeds, at the same Sonnet prices. Sonnet is particularly good at:
- Coding: Autonomously writes, edits, and runs code with reasoning and troubleshooting
- Data science: Augments human data science expertise; navigates unstructured data while using multiple tools for insights
- Visual processing: excelling at interpreting charts, graphs, and images, accurately transcribing text to derive insights beyond just the text alone
- Agentic tasks: exceptional tool use, making it great at agentic tasks (i.e. complex, multi-step problem solving tasks that require engaging with other systems)
For the latest version (2024-10-23), check out [Claude 3.5 Sonnet](/anthropic/claude-3.5-sonnet).
New Claude 3.5 Sonnet delivers better-than-Opus capabilities, faster-than-Sonnet speeds, at the same Sonnet prices. Sonnet is particularly good at:
- Coding: Scores ~49% on SWE-Bench Verified, higher than the last best score, and without any fancy prompt scaffolding
- Data science: Augments human data science expertise; navigates unstructured data while using multiple tools for insights
- Visual processing: excelling at interpreting charts, graphs, and images, accurately transcribing text to derive insights beyond just the text alone
- Agentic tasks: exceptional tool use, making it great at agentic tasks (i.e. complex, multi-step problem solving tasks that require engaging with other systems)
New Claude 3.5 Sonnet delivers better-than-Opus capabilities, faster-than-Sonnet speeds, at the same Sonnet prices. Sonnet is particularly good at:
- Coding: Scores ~49% on SWE-Bench Verified, higher than the last best score, and without any fancy prompt scaffolding
- Data science: Augments human data science expertise; navigates unstructured data while using multiple tools for insights
- Visual processing: excelling at interpreting charts, graphs, and images, accurately transcribing text to derive insights beyond just the text alone
- Agentic tasks: exceptional tool use, making it great at agentic tasks (i.e. complex, multi-step problem solving tasks that require engaging with other systems)
description:"Claude 3.7 Sonnet is an advanced large language model with improved reasoning, coding, and problem-solving capabilities. It introduces a hybrid reasoning approach, allowing users to choose between rapid responses and extended, step-by-step processing for complex tasks. The model demonstrates notable improvements in coding, particularly in front-end development and full-stack updates, and excels in agentic workflows, where it can autonomously navigate multi-step processes. \n\nClaude 3.7 Sonnet maintains performance parity with its predecessor in standard mode while offering an extended reasoning mode for enhanced accuracy in math, coding, and instruction-following tasks.\n\nRead more at the [blog post here](https://www.anthropic.com/news/claude-3-7-sonnet)"
description:"Claude 3.7 Sonnet is an advanced large language model with improved reasoning, coding, and problem-solving capabilities. It introduces a hybrid reasoning approach, allowing users to choose between rapid responses and extended, step-by-step processing for complex tasks. The model demonstrates notable improvements in coding, particularly in front-end development and full-stack updates, and excels in agentic workflows, where it can autonomously navigate multi-step processes. \n\nClaude 3.7 Sonnet maintains performance parity with its predecessor in standard mode while offering an extended reasoning mode for enhanced accuracy in math, coding, and instruction-following tasks.\n\nRead more at the [blog post here](https://www.anthropic.com/news/claude-3-7-sonnet)"
description:"Claude 3.7 Sonnet is an advanced large language model with improved reasoning, coding, and problem-solving capabilities. It introduces a hybrid reasoning approach, allowing users to choose between rapid responses and extended, step-by-step processing for complex tasks. The model demonstrates notable improvements in coding, particularly in front-end development and full-stack updates, and excels in agentic workflows, where it can autonomously navigate multi-step processes. \n\nClaude 3.7 Sonnet maintains performance parity with its predecessor in standard mode while offering an extended reasoning mode for enhanced accuracy in math, coding, and instruction-following tasks.\n\nRead more at the [blog post here](https://www.anthropic.com/news/claude-3-7-sonnet)"
description:"Claude Opus 4 is benchmarked as the world’s best coding model, at time of release, bringing sustained performance on complex, long-running tasks and agent workflows. It sets new benchmarks in software engineering, achieving leading results on SWE-bench (72.5%) and Terminal-bench (43.2%). Opus 4 supports extended, agentic workflows, handling thousands of task steps continuously for hours without degradation. \n\nRead more at the [blog post here](https://www.anthropic.com/news/claude-4)"
Claude Sonnet 4 significantly enhances the capabilities of its predecessor, Sonnet 3.7, excelling in both coding and reasoning tasks with improved precision and controllability. Achieving state-of-the-art performance on SWE-bench (72.7%), Sonnet 4 balances capability and computational efficiency, making it suitable for a broad range of applications from routine coding tasks to complex software development projects. Key enhancements include improved autonomous codebase navigation, reduced error rates in agent-driven workflows, and increased reliability in following intricate instructions. Sonnet 4 is optimized for practical everyday use, providing advanced reasoning capabilities while maintaining efficiency and responsiveness in diverse internal and external scenarios.
Read more at the [blog post here](https://www.anthropic.com/news/claude-4)
description:'Arcee Blitz is a 24B‑parameter dense model distilled from DeepSeek and built on Mistral architecture for "everyday" chat. The distillation‑plus‑refinement pipeline trims compute while keeping DeepSeek‑style reasoning, so Blitz punches above its weight on MMLU, GSM‑8K and BBH compared with other mid‑size open models. With a default 128k context window and competitive throughput, it serves as a cost‑efficient workhorse for summarization, brainstorming and light code help. Internally, Arcee uses Blitz as the default writer in Conductor pipelines when the heavier Virtuoso line is not required. Users therefore get near‑70B quality at ~⅓ the latency and price. '
description:'Caller Large is Arcee''s specialist "function‑calling" SLM built to orchestrate external tools and APIs. Instead of maximizing next‑token accuracy, training focuses on structured JSON outputs, parameter extraction and multi‑step tool chains, making Caller a natural choice for retrieval‑augmented generation, robotic process automation or data‑pull chatbots. It incorporates a routing head that decides when (and how) to invoke a tool versus answering directly, reducing hallucinated calls. The model is already the backbone of Arcee Conductor''s auto‑tool mode, where it parses user intent, emits clean function signatures and hands control back once the tool response is ready. Developers thus gain an OpenAI‑style function‑calling UX without handing requests to a frontier‑scale model. '
description:'Coder‑Large is a 32B‑parameter offspring of Qwen2.5‑Instruct that has been further trained on permissively‑licensed GitHub, CodeSearchNet and synthetic bug‑fix corpora. It supports a 32k context window, enabling multi‑file refactoring or long diff review in a single call, and understands 30‑plus programming languages with special attention to TypeScript, Go and Terraform. Internal benchmarks show 5–8pt gains over CodeLlama‑34B‑Python on HumanEval and competitive BugFix scores thanks to a reinforcement pass that rewards compilable output. The model emits structured explanations alongside code blocks by default, making it suitable for educational tooling as well as production copilot scenarios. Cost‑wise, Together AI prices it well below proprietary incumbents, so teams can scale interactive coding without runaway spend. '
description: 'Maestro Reasoning is Arcee''s flagship analysis model:a 32B‑parameter derivative of Qwen2.5‑32B tuned with DPO and chain‑of‑thought RL for step‑by‑step logic. Compared to the earlier 7B preview, the production 32B release widens the context window to 128k tokens and doubles pass‑rate on MATH and GSM‑8K, while also lifting code completion accuracy. Its instruction style encourages structured "thought → answer" traces that can be parsed or hidden according to user preference. That transparency pairs well with audit‑focused industries like finance or healthcare where seeing the reasoning path matters. In Arcee Conductor, Maestro is automatically selected for complex, multi‑constraint queries that smaller SLMs bounce. '
description:'Spotlight is a 7‑billion‑parameter vision‑language model derived from Qwen2.5‑VL and fine‑tuned by Arcee AI for tight image‑text grounding tasks. It offers a 32k‑token context window, enabling rich multimodal conversations that combine lengthy documents with one or more images. Training emphasized fast inference on consumer GPUs while retaining strong captioning, visual‐question‑answering, and diagram‑analysis accuracy. As a result, Spotlight slots neatly into agent workflows where screenshots, charts or UI mock‑ups need to be interpreted on the fly. Early benchmarks show it matching or out‑scoring larger VLMs such as LLaVA‑1.6 13B on popular VQA and POPE alignment tests. '
description:Virtuoso‑Large is Arcee's top‑tier general‑purpose LLM at 72B parameters, tuned to tackle cross‑domain reasoning, creative writing and enterprise QA. Unlike many 70B peers, it retains the 128k context inherited from Qwen2.5, letting it ingest books, codebases or financial filings wholesale. Training blended DeepSeekR1 distillation, multi‑epoch supervised fine‑tuning and a final DPO/RLHF alignment stage, yielding strong performance on BIG‑Bench‑Hard, GSM‑8K and long‑context Needle‑In‑Haystack tests. Enterprises use Virtuoso‑Large as the "fallback" brain in Conductor pipelines when other SLMs flag low confidence. Despite its size, aggressive KV‑cache optimizations keep first‑token latency in the low‑second range on 8×H100 nodes, making it a practical production‑grade powerhouse.
description:'Virtuoso‑Medium‑v2 is a 32B model distilled from DeepSeek‑v3 logits and merged back onto a Qwen2.5 backbone, yielding a sharper, more factual successor to the original Virtuoso Medium. The team harvested ~1.1B logit tokens and applied "fusion‑merging" plus DPO alignment, which pushed scores past Arcee‑Nova2024 and many 40B‑plus peers on MMLU‑Pro, MATH and HumanEval. With a 128k context and aggressive quantization options (from BF16 down to 4‑bit GGUF), it balances capability with deployability on single‑GPU nodes. Typical use cases include enterprise chat assistants, technical writing aids and medium‑complexity code drafting where Virtuoso‑Large would be overkill. '
Dolphin 2.9 is designed for instruction following, conversational, and coding. This model is a finetune of [Mixtral 8x22B Instruct](/models/mistralai/mixtral-8x22b-instruct). It features a 64k context length and was fine-tuned with a 16k sequence length using ChatML templates.
This model is a successor to [Dolphin Mixtral 8x7B](/models/cognitivecomputations/dolphin-mixtral-8x7b).
The model is uncensored and is stripped of alignment and bias. It requires an external alignment layer for ethical use. Users are cautioned to use this highly compliant model responsibly, as detailed in a blog post about uncensored models at [erichartford.com/uncensored-models](https://erichartford.com/uncensored-models).
Command A is an open-weights 111B parameter model with a 256k context window focused on delivering great performance across agentic, multilingual, and coding use cases.
Compared to other leading proprietary and open-weights models Command A delivers maximum performance with minimum hardware costs, excelling on business-critical agentic and multilingual tasks.
Command-R is a 35B parameter model that performs conversational language tasks at a higher quality, more reliably, and with a longer context than previous models. It can be used for complex workflows like code generation, retrieval augmented generation (RAG), tool use, and agents.
Read the launch post [here](https://txt.cohere.com/command-r/).
Use of this model is subject to Cohere's [Usage Policy](https://docs.cohere.com/docs/usage-policy) and [SaaS Agreement](https://cohere.com/saas-agreement).
command-r-08-2024 is an update of the [Command R](/models/cohere/command-r) with improved performance for multilingual retrieval-augmented generation (RAG) and tool use. More broadly, it is better at math, code and reasoning and is competitive with the previous version of the larger Command R+ model.
Read the launch post [here](https://docs.cohere.com/changelog/command-gets-refreshed).
Use of this model is subject to Cohere's [Usage Policy](https://docs.cohere.com/docs/usage-policy) and [SaaS Agreement](https://cohere.com/saas-agreement).
Command R+ is a new, 104B-parameter LLM from Cohere. It's useful for roleplay, general consumer usecases, and Retrieval Augmented Generation (RAG).
It offers multilingual support for ten key languages to facilitate global business operations. See benchmarks and the launch post [here](https://txt.cohere.com/command-r-plus-microsoft-azure/).
Use of this model is subject to Cohere's [Usage Policy](https://docs.cohere.com/docs/usage-policy) and [SaaS Agreement](https://cohere.com/saas-agreement).
command-r-plus-08-2024 is an update of the [Command R+](/models/cohere/command-r-plus) with roughly 50% higher throughput and 25% lower latencies as compared to the previous Command R+ version, while keeping the hardware footprint the same.
Read the launch post [here](https://docs.cohere.com/changelog/command-gets-refreshed).
Use of this model is subject to Cohere's [Usage Policy](https://docs.cohere.com/docs/usage-policy) and [SaaS Agreement](https://cohere.com/saas-agreement).
Command R+ is a new, 104B-parameter LLM from Cohere. It's useful for roleplay, general consumer usecases, and Retrieval Augmented Generation (RAG).
It offers multilingual support for ten key languages to facilitate global business operations. See benchmarks and the launch post [here](https://txt.cohere.com/command-r-plus-microsoft-azure/).
Use of this model is subject to Cohere's [Usage Policy](https://docs.cohere.com/docs/usage-policy) and [SaaS Agreement](https://cohere.com/saas-agreement).
Command-R is a 35B parameter model that performs conversational language tasks at a higher quality, more reliably, and with a longer context than previous models. It can be used for complex workflows like code generation, retrieval augmented generation (RAG), tool use, and agents.
Read the launch post [here](https://txt.cohere.com/command-r/).
Use of this model is subject to Cohere's [Usage Policy](https://docs.cohere.com/docs/usage-policy) and [SaaS Agreement](https://cohere.com/saas-agreement).
Command R7B (12-2024) is a small, fast update of the Command R+ model, delivered in December 2024. It excels at RAG, tool use, agents, and similar tasks requiring complex reasoning and multiple steps.
Use of this model is subject to Cohere's [Usage Policy](https://docs.cohere.com/docs/usage-policy) and [SaaS Agreement](https://cohere.com/saas-agreement).
Command is an instruction-following conversational model that performs language tasks with high quality, more reliably and with a longer context than our base generative models.
Use of this model is subject to Cohere's [Usage Policy](https://docs.cohere.com/docs/usage-policy) and [SaaS Agreement](https://cohere.com/saas-agreement).
DeepSeek-V3 is the latest model from the DeepSeek team, building upon the instruction following and coding abilities of the previous versions. Pre-trained on nearly 15 trillion tokens, the reported evaluations reveal that the model outperforms other open-source models and rivals leading closed-source models.
For model details, please visit [the DeepSeek-V3 repo](https://github.com/deepseek-ai/DeepSeek-V3) for more information, or see the [launch announcement](https://api-docs.deepseek.com/news/news1226).
description:DeepSeek Prover V2 is a 671B parameter model, speculated to be geared towards logic and mathematics. Likely an upgrade from [DeepSeek-Prover-V1.5](https://huggingface.co/deepseek-ai/DeepSeek-Prover-V1.5-RL) Not much is known about the model yet, as DeepSeek released it on Hugging Face without an announcement or description.
context_length:131072
architecture:
modality:text->text
input_modalities:
- text
output_modalities:
- text
tokenizer:DeepSeek
instruct_type:null
pricing:
prompt:'0.0000005'
completion:'0.00000218'
input_cache_read:''
input_cache_write:''
request:'0'
image:'0'
web_search:'0'
internal_reasoning:'0'
unit:1
currency:USD
supported_parameters:
- max_tokens
- temperature
- top_p
- stop
- frequency_penalty
- presence_penalty
- seed
- top_k
- min_p
- repetition_penalty
- logit_bias
- response_format
model_provider:deepseek
Some files were not shown because too many files have changed in this diff
Show More
Reference in New Issue
Block a user
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
