chore(deps): update ghcr.io/open-webui/open-webui docker tag to v0.8.8

[renovate]

ℹ️ Note

This PR body was truncated due to platform limits.

This PR contains the following updates:

Package	Update	Change
ghcr.io/open-webui/open-webui	minor	0.6.41 → v0.8.8

---

Release Notes

open-webui/open-webui (ghcr.io/open-webui/open-webui)

v0.8.8

Compare Source

[0.8.8] - 2026-03-02

Added

• 📁 Open Terminal file moving. Users can now move files and folders between directories in the Open Terminal file browser by dragging and dropping them. Commit, Commit, Commit, Commit
• 📄 Open Terminal HTML file preview. Users can now preview HTML files directly in the Open Terminal file browser, with a rendered iframe view and source toggle, enabling iterative AI editing of HTML files. Commit, Commit
• 🌐 Open Terminal WebSocket proxy. Added a new WebSocket proxy endpoint for interactive terminal sessions, enabling real-time bidirectional terminal communication with the terminal server. Commit
• ⚙️ Open Terminal feature toggle. Administrators can now enable or disable the Interactive Terminal feature for Open Terminal via configuration on the terminal server, controlling access to terminal routes. Commit
• 🔄 General improvements. Various improvements were implemented across the application to enhance performance, stability, and security.
• 🌐 Translations for Simplified Chinese, Traditional Chinese, Irish, and Catalan were enhanced and expanded.

Fixed

• 🔧 Middleware variable shadowing. Fixed a variable shadowing issue in the middleware that could cause incorrect tool output processing during chat. #​22145
• ⚡ ChatControls reactivity fix. Fixed a Svelte reactivity issue where the active tab state in the ChatControls panel was not properly saved when switching between chats. #​22127
• 🔧 ChatControls TypeScript fix. Fixed a TypeScript syntax error in ChatControls.svelte where the module script block was missing lang="ts", causing esbuild to fail during vite dev. #​22131
• 🔌 Open Terminal tools for direct connections. Fixed an issue where Open Terminal tools were not available to the model when the terminal was configured via direct connection settings, ensuring users can now interact with terminal files and operations through the AI. #​22137
• 📜 Chat history pagination. Fixed an issue where older messages in long chats were not loaded when scrolling to the top. Commit, Commit
• 🔧 Terminal tool null parameter handling. Fixed a bug where null parameters in terminal tool calls were sent as the string "None" instead of being omitted, causing 422 validation errors from the open-terminal server. #​22124, #​22144

Changed

v0.8.8

Compare Source

Added

• 📁 Open Terminal file moving. Users can now move files and folders between directories in the Open Terminal file browser by dragging and dropping them. Commit, Commit, Commit, Commit
• 📄 Open Terminal HTML file preview. Users can now preview HTML files directly in the Open Terminal file browser, with a rendered iframe view and source toggle, enabling iterative AI editing of HTML files. Commit, Commit
• 🌐 Open Terminal WebSocket proxy. Added a new WebSocket proxy endpoint for interactive terminal sessions, enabling real-time bidirectional terminal communication with the terminal server. Commit
• ⚙️ Open Terminal feature toggle. Administrators can now enable or disable the Interactive Terminal feature for Open Terminal via configuration on the terminal server, controlling access to terminal routes. Commit
• 🔄 General improvements. Various improvements were implemented across the application to enhance performance, stability, and security.
• 🌐 Translations for Simplified Chinese, Traditional Chinese, Irish, and Catalan were enhanced and expanded.

Fixed

• 🔧 Middleware variable shadowing. Fixed a variable shadowing issue in the middleware that could cause incorrect tool output processing during chat. #​22145
• ⚡ ChatControls reactivity fix. Fixed a Svelte reactivity issue where the active tab state in the ChatControls panel was not properly saved when switching between chats. #​22127
• 🔧 ChatControls TypeScript fix. Fixed a TypeScript syntax error in ChatControls.svelte where the module script block was missing lang="ts", causing esbuild to fail during vite dev. #​22131
• 🔌 Open Terminal tools for direct connections. Fixed an issue where Open Terminal tools were not available to the model when the terminal was configured via direct connection settings, ensuring users can now interact with terminal files and operations through the AI. #​22137
• 📜 Chat history pagination. Fixed an issue where older messages in long chats were not loaded when scrolling to the top. Commit, Commit
• 🔧 Terminal tool null parameter handling. Fixed a bug where null parameters in terminal tool calls were sent as the string "None" instead of being omitted, causing 422 validation errors from the open-terminal server. #​22124, #​22144

Changed

v0.8.7

Compare Source

[0.8.7] - 2026-03-01

Fixed

• 🔒 Connection access control privacy. Tool server and terminal connections without explicit access grants are now private (admin-only) by default, fixing a bug where connections configured with no access grants were visible to all users instead of being restricted. Commit
• 🧠 ChatControls memory leak. The ChatControls panel no longer leaks event listeners, ResizeObserver instances, and media query handlers when navigating between chats, fixing memory accumulation that could degrade performance during extended use. #​22112
• 💾 Temporary chat params preservation. Model parameters are now correctly saved when creating a temporary chat, ensuring custom settings like temperature and top_p persist across the session. Commit
• ⚡ Faster artifact content updates. Artifact content extraction during streaming is now debounced via requestAnimationFrame, reducing redundant DOM reads and improving CPU efficiency when tokens arrive faster than the browser can paint. Commit

v0.8.7

Compare Source

Fixed

• 🔒 Connection access control privacy. Tool server and terminal connections without explicit access grants are now private (admin-only) by default, fixing a bug where connections configured with no access grants were visible to all users instead of being restricted. Commit
• 🧠 ChatControls memory leak. The ChatControls panel no longer leaks event listeners, ResizeObserver instances, and media query handlers when navigating between chats, fixing memory accumulation that could degrade performance during extended use. #​22112
• 💾 Temporary chat params preservation. Model parameters are now correctly saved when creating a temporary chat, ensuring custom settings like temperature and top_p persist across the session. Commit
• ⚡ Faster artifact content updates. Artifact content extraction during streaming is now debounced via requestAnimationFrame, reducing redundant DOM reads and improving CPU efficiency when tokens arrive faster than the browser can paint. Commit

v0.8.6

Compare Source

[0.8.6] - 2026-03-01

Added

• 🖥️ Open Terminal integration. Users can now connect to Open Terminal instances to browse, read, and upload files directly in chat, with the terminal acting as an always-on tool. File navigation includes folder browsing, image and PDF previews, drag-and-drop uploads, directory creation, and file deletion. The current working directory is automatically injected into tool descriptions for context-aware commands. Commit, Commit, Commit
• 📄 Terminal file creation. Users can now create new empty files directly in the Open Terminal file browser, in addition to the existing folder creation functionality. Commit
• ✏️ Terminal file editing. Users can now edit text files directly in the Open Terminal file browser, with the ability to save changes back to the terminal. Commit
• 🛠️ Terminal file preview toolbar. The Open Terminal file browser now displays contextual toolbar buttons based on file type, including preview/source toggle for Markdown and CSV files, reset view for images, and improved editing controls for text files. Commit
• 🔄 Terminal file write refresh. The file browser now automatically refreshes when files are written or modified via the write_file or replace_file_content tools, eliminating the need to manually refresh. Commit
• 🛡️ Docker image SBOM attestation. Docker images now include a Software Bill of Materials (SBOM) for vulnerability scanning and supply chain security compliance. #​21779, Commit
• 📡 Reporting-Endpoints security header. Administrators can now configure a Reporting-Endpoints header via the REPORTING_ENDPOINTS environment variable to receive CSP violation reports directly, aiding in security policy debugging and hardening. #​21830
• 🎯 Action button priority sorting. Action buttons under assistant messages now appear in a consistent order based on the priority field from function Valves, allowing developers to control button placement. #​21790
• 🏷️ Public/Private model filtering. The Admin Settings Model listing now displays Public/Private badges and includes filter options to easily view public or private models. #​21732, #​21797
• 👁️ Show/Hide all models bulk action. Administrators can now show or hide all models at once from the Admin Settings Models page Actions menu, making it faster to manage model visibility. Bulk actions now display a single toast notification on success for better user feedback. #​21838, #​21958
• 🔐 Individual user sharing control. Administrators can now disable individual user sharing via the USER_PERMISSIONS_ACCESS_GRANTS_ALLOW_USERS environment variable, allowing only group-based sharing when set to false. #​21793, Commit, Commit
• 🔄 OAuth profile sync on login. Administrators can now enable automatic synchronization of user profile name and email from OAuth providers on login via the OAUTH_UPDATE_NAME_ON_LOGIN and OAUTH_UPDATE_EMAIL_ON_LOGIN environment variables. #​21787, Commit
• 👥 Default group share permission. Administrators can now configure the default sharing permission for new groups via the DEFAULT_GROUP_SHARE_PERMISSION environment variable, controlling whether anyone, no one, or only members can share to new groups. Commit
• 💨 Streaming performance. Chat responses now render more efficiently during streaming, reducing CPU usage and improving responsiveness. Commit
• 🧮 Streaming message comparison. Chat message updates during streaming are now faster thanks to an optimization that skips expensive comparisons when content changes. #​21884
• 🚀 Streaming scroll optimization. Chat auto-scroll during streaming is now more efficient by batching scroll operations via requestAnimationFrame, reducing unnecessary layout reflows when tokens arrive faster than the browser can paint. #​21946
• 📋 Message cloning performance. Chat message cloning during streaming is now more efficient thanks to the use of structuredClone() instead of JSON.parse(JSON.stringify(...)). #​21948
• 🎯 Faster code block rendering. Chat message updates during streaming are now faster. #​22101
• 📊 Faster status history display. Chat message updates during streaming are now faster. #​22103
• 🛠️ Faster tool result handling. Tool execution results are now handled more efficiently, improving streaming performance. #​22104
• 💾 Faster model and file operations. Model selection, file preparation, and history saving are now faster. #​22102
• 🛠️ Tool server advanced options toggle. Advanced OpenAPI configuration options in the tool server modal are now hidden by default behind a toggle, simplifying the interface for basic setups. The admin settings tab was also renamed from "Tools" to "Integrations" for clearer organization. Commit, Commit
• 🔧 Faster tool loading. Tool access control now skips an unnecessary database query when no tools are attached to the request, slightly improving performance. #​21873
• ➗ Faster math rendering. Mathematical notation now renders more efficiently, improving responsiveness when displaying equations in chat. #​21880
• 🏎️ Faster message list updates. The chat message list now rebuilds at most once per animation frame during streaming, reducing CPU overhead. #​21885
• 📋 Faster message rendering. Chat message rendering is now more efficient during streaming. #​22086
• 🗄️ Faster real-time chat updates. Chat responses now process faster with improved handling for concurrent users. #​22087
• 📝 Faster status persistence. Only final status updates are now saved to the database during streaming, reducing unnecessary writes. #​22085
• 🔄 Faster event matching. Event handling in the socket handler is now more efficient. Commit
• 🔀 General improvements. Various improvements were implemented across the application to enhance performance, stability, and security.
• 🌐 Translation updates. Translations for German, Portuguese (Brazil), Simplified Chinese, Traditional Chinese, Catalan, and Spanish were enhanced and expanded.

Fixed

• 🗄️ Database migration execution. Database migrations now run correctly on startup, fixing a circular import issue that caused schema updates to fail silently. #​21848, Commit
• 🔔 Notification HTML escaping. Notification messages now properly escape HTML content, matching the behavior in chat messages and ensuring consistent rendering across the interface. #​21860, Commit
• 🛠️ Tool call JSON error handling. Chat no longer crashes when models generate malformed JSON in tool call arguments; instead, a descriptive error message is returned to the model for retry. #​21984, Commit
• 🧠 Reasoning model KV cache preservation. Reasoning model thinking tags are no longer stored as HTML in the database, preserving KV cache efficiency for backends like llama.cpp and ensuring faster subsequent conversation turns. #​21815, Commit
• ⚡ Duplicate model execution prevention. Models are no longer called twice when no tools are configured, eliminating unnecessary API requests and reducing latency. #​21802, Commit
• 🔐 OAuth session database error. OAuth login no longer fails with a database error when creating sessions, fixing the "'NoneType' object has no attribute 'id'" and "can't adapt type 'dict'" errors that occurred during OAuth group creation. #​21788
• 👤 User sharing permission enforcement. The user sharing option now correctly respects the USER_PERMISSIONS_ACCESS_GRANTS_ALLOW_USERS setting, fixing an issue where sharing to individual users was incorrectly allowed even when disabled. #​21856, Commit, Commit
• 🔑 Password manager autofill. Password manager autofill (like iCloud Passwords, 1Password, Bitwarden) now correctly captures filled-in passwords, fixing login failures where the password appeared filled but was sent as empty. #​21869, Commit
• 📝 RAG template duplication. RAG templates are no longer duplicated in chat messages when models make multiple tool calls, preventing hallucinations and incorrect tool usage. #​21780, Commit
• 📋 Audit log stdout. Audit logs now correctly appear on stdout when the ENABLE_AUDIT_STDOUT environment variable is set to true, aligning runtime behavior with the intended configuration. #​21777
• 🎯 Function valve priority resolution. Function priorities defined in code are now correctly applied when no custom value has been saved in the database, ensuring consistent action button and filter ordering. #​21841
• 📄 Web content knowledge base append. Processing web URLs with overwrite=false now correctly appends content to existing knowledge bases instead of silently doing nothing, fixing a regression where no content was being added. #​21786, Commit
• 🔍 Web search domain filter config. The WEB_SEARCH_DOMAIN_FILTER_LIST environment variable is now correctly read and applied, fixing an issue where domain filtering for web searches always used an empty default value. #​21964, #​20186
• 🧹 Tooltip memory leak. Tooltip instances are now properly destroyed when elements change, fixing a memory leak that could cause performance issues over time. #​21969
• ⌨️ MessageInput memory leak. Event listeners in the message input component are now properly cleaned up, preventing a memory leak that could cause page crashes during extended use. #​21968
• 📝 Notes memory leak. Event listeners in the Notes component are now properly cleaned up, fixing a memory leak that could cause page crashes during extended use. #​21963
• 🏗️ Model create memory leak. Event listeners in the model creation page are now properly cleaned up, fixing a memory leak that could cause page crashes during extended use. #​21966
• 💬 MentionList memory leak. Event listeners in the MentionList component are now properly cleaned up, fixing a memory leak that could cause page crashes during extended use. #​21965
• 📐 Sidebar memory leak. Event listeners in the Sidebar component are now properly cleaned up, fixing a memory leak that could cause page crashes during extended use. #​22082
• 🎨 Sidebar user menu positioning. The sidebar user menu no longer drifts rightward when the sidebar is resized, keeping the menu properly aligned with its trigger. #​21853
• 💻 Code block UI. Code block headers are now sticky and properly positioned, with language labels now showing tooltips for truncated text. Commit
• 📊 Multi-model responses horizontal scroll. The model list in multi-model responses tabs now has horizontal scroll support, making all models accessible on desktop screens. #​21800, Commit
• 🎭 TailwindCSS gray color theme. Custom gray color palette is now correctly applied to the CSS root theme layer, fixing an issue where --color-gray-x variables were missing. #​21900, #​21899
• 📎 Broken documentation links. Fixed broken links in the backend config and admin settings that pointed to outdated documentation locations. #​21904
• 🔓 OAuth session token decryption. OAuth sessions are now properly detached from the database context before token decryption, preventing potential database session conflicts when reading encrypted tokens. #​21794
• 🕐 Chat timestamp i18n fix. Chat timestamps in the sidebar now display correctly, fixing an issue where the time ago format (e.g., "5m", "2h", "3d") was not being localized properly due to incorrect variable casing in the translation function. Commit
• 🍞 Model toast notification fix. Hiding or showing a single model now displays only one toast notification instead of two, removing the redundant generic "model updated" message when a specific action toast is shown. #​22079
• 📡 Offline mode embedding model fix. Open WebUI no longer attempts to download embedding models when in offline mode, fixing error logs that occurred when trying to fetch models that weren't cached locally. #​22106, #​21405

v0.8.6

Compare Source

Added

• 🖥️ Open Terminal integration. Users can now connect to Open Terminal instances to browse, read, and upload files directly in chat, with the terminal acting as an always-on tool. File navigation includes folder browsing, image and PDF previews, drag-and-drop uploads, directory creation, and file deletion. The current working directory is automatically injected into tool descriptions for context-aware commands. Commit, Commit, Commit
• 📄 Terminal file creation. Users can now create new empty files directly in the Open Terminal file browser, in addition to the existing folder creation functionality. Commit
• ✏️ Terminal file editing. Users can now edit text files directly in the Open Terminal file browser, with the ability to save changes back to the terminal. Commit
• 🛠️ Terminal file preview toolbar. The Open Terminal file browser now displays contextual toolbar buttons based on file type, including preview/source toggle for Markdown and CSV files, reset view for images, and improved editing controls for text files. Commit
• 🔄 Terminal file write refresh. The file browser now automatically refreshes when files are written or modified via the write_file or replace_file_content tools, eliminating the need to manually refresh. Commit
• 🛡️ Docker image SBOM attestation. Docker images now include a Software Bill of Materials (SBOM) for vulnerability scanning and supply chain security compliance. #​21779, Commit
• 📡 Reporting-Endpoints security header. Administrators can now configure a Reporting-Endpoints header via the REPORTING_ENDPOINTS environment variable to receive CSP violation reports directly, aiding in security policy debugging and hardening. #​21830
• 🎯 Action button priority sorting. Action buttons under assistant messages now appear in a consistent order based on the priority field from function Valves, allowing developers to control button placement. #​21790
• 🏷️ Public/Private model filtering. The Admin Settings Model listing now displays Public/Private badges and includes filter options to easily view public or private models. #​21732, #​21797
• 👁️ Show/Hide all models bulk action. Administrators can now show or hide all models at once from the Admin Settings Models page Actions menu, making it faster to manage model visibility. Bulk actions now display a single toast notification on success for better user feedback. #​21838, #​21958
• 🔐 Individual user sharing control. Administrators can now disable individual user sharing via the USER_PERMISSIONS_ACCESS_GRANTS_ALLOW_USERS environment variable, allowing only group-based sharing when set to false. #​21793, Commit, Commit
• 🔄 OAuth profile sync on login. Administrators can now enable automatic synchronization of user profile name and email from OAuth providers on login via the OAUTH_UPDATE_NAME_ON_LOGIN and OAUTH_UPDATE_EMAIL_ON_LOGIN environment variables. #​21787, Commit
• 👥 Default group share permission. Administrators can now configure the default sharing permission for new groups via the DEFAULT_GROUP_SHARE_PERMISSION environment variable, controlling whether anyone, no one, or only members can share to new groups. Commit
• 💨 Streaming performance. Chat responses now render more efficiently during streaming, reducing CPU usage and improving responsiveness. Commit
• 🧮 Streaming message comparison. Chat message updates during streaming are now faster thanks to an optimization that skips expensive comparisons when content changes. #​21884
• 🚀 Streaming scroll optimization. Chat auto-scroll during streaming is now more efficient by batching scroll operations via requestAnimationFrame, reducing unnecessary layout reflows when tokens arrive faster than the browser can paint. #​21946
• 📋 Message cloning performance. Chat message cloning during streaming is now more efficient thanks to the use of structuredClone() instead of JSON.parse(JSON.stringify(...)). #​21948
• 🎯 Faster code block rendering. Chat message updates during streaming are now faster. #​22101
• 📊 Faster status history display. Chat message updates during streaming are now faster. #​22103
• 🛠️ Faster tool result handling. Tool execution results are now handled more efficiently, improving streaming performance. #​22104
• 💾 Faster model and file operations. Model selection, file preparation, and history saving are now faster. #​22102
• 🛠️ Tool server advanced options toggle. Advanced OpenAPI configuration options in the tool server modal are now hidden by default behind a toggle, simplifying the interface for basic setups. The admin settings tab was also renamed from "Tools" to "Integrations" for clearer organization. Commit, Commit
• 🔧 Faster tool loading. Tool access control now skips an unnecessary database query when no tools are attached to the request, slightly improving performance. #​21873
• ➗ Faster math rendering. Mathematical notation now renders more efficiently, improving responsiveness when displaying equations in chat. #​21880
• 🏎️ Faster message list updates. The chat message list now rebuilds at most once per animation frame during streaming, reducing CPU overhead. #​21885
• 📋 Faster message rendering. Chat message rendering is now more efficient during streaming. #​22086
• 🗄️ Faster real-time chat updates. Chat responses now process faster with improved handling for concurrent users. #​22087
• 📝 Faster status persistence. Only final status updates are now saved to the database during streaming, reducing unnecessary writes. #​22085
• 🔄 Faster event matching. Event handling in the socket handler is now more efficient. Commit
• 🔀 General improvements. Various improvements were implemented across the application to enhance performance, stability, and security.
• 🌐 Translation updates. Translations for German, Portuguese (Brazil), Simplified Chinese, Traditional Chinese, Catalan, and Spanish were enhanced and expanded.

Fixed

• 🗄️ Database migration execution. Database migrations now run correctly on startup, fixing a circular import issue that caused schema updates to fail silently. #​21848, Commit
• 🔔 Notification HTML escaping. Notification messages now properly escape HTML content, matching the behavior in chat messages and ensuring consistent rendering across the interface. #​21860, Commit
• 🛠️ Tool call JSON error handling. Chat no longer crashes when models generate malformed JSON in tool call arguments; instead, a descriptive error message is returned to the model for retry. #​21984, Commit
• 🧠 Reasoning model KV cache preservation. Reasoning model thinking tags are no longer stored as HTML in the database, preserving KV cache efficiency for backends like llama.cpp and ensuring faster subsequent conversation turns. #​21815, Commit
• ⚡ Duplicate model execution prevention. Models are no longer called twice when no tools are configured, eliminating unnecessary API requests and reducing latency. #​21802, Commit
• 🔐 OAuth session database error. OAuth login no longer fails with a database error when creating sessions, fixing the "'NoneType' object has no attribute 'id'" and "can't adapt type 'dict'" errors that occurred during OAuth group creation. #​21788
• 👤 User sharing permission enforcement. The user sharing option now correctly respects the USER_PERMISSIONS_ACCESS_GRANTS_ALLOW_USERS setting, fixing an issue where sharing to individual users was incorrectly allowed even when disabled. #​21856, Commit, Commit
• 🔑 Password manager autofill. Password manager autofill (like iCloud Passwords, 1Password, Bitwarden) now correctly captures filled-in passwords, fixing login failures where the password appeared filled but was sent as empty. #​21869, Commit
• 📝 RAG template duplication. RAG templates are no longer duplicated in chat messages when models make multiple tool calls, preventing hallucinations and incorrect tool usage. #​21780, Commit
• 📋 Audit log stdout. Audit logs now correctly appear on stdout when the ENABLE_AUDIT_STDOUT environment variable is set to true, aligning runtime behavior with the intended configuration. #​21777
• 🎯 Function valve priority resolution. Function priorities defined in code are now correctly applied when no custom value has been saved in the database, ensuring consistent action button and filter ordering. #​21841
• 📄 Web content knowledge base append. Processing web URLs with overwrite=false now correctly appends content to existing knowledge bases instead of silently doing nothing, fixing a regression where no content was being added. #​21786, Commit
• 🔍 Web search domain filter config. The WEB_SEARCH_DOMAIN_FILTER_LIST environment variable is now correctly read and applied, fixing an issue where domain filtering for web searches always used an empty default value. #​21964, #​20186
• 🧹 Tooltip memory leak. Tooltip instances are now properly destroyed when elements change, fixing a memory leak that could cause performance issues over time. #​21969
• ⌨️ MessageInput memory leak. Event listeners in the message input component are now properly cleaned up, preventing a memory leak that could cause page crashes during extended use. #​21968
• 📝 Notes memory leak. Event listeners in the Notes component are now properly cleaned up, fixing a memory leak that could cause page crashes during extended use. #​21963
• 🏗️ Model create memory leak. Event listeners in the model creation page are now properly cleaned up, fixing a memory leak that could cause page crashes during extended use. #​21966
• 💬 MentionList memory leak. Event listeners in the MentionList component are now properly cleaned up, fixing a memory leak that could cause page crashes during extended use. #​21965
• 📐 Sidebar memory leak. Event listeners in the Sidebar component are now properly cleaned up, fixing a memory leak that could cause page crashes during extended use. #​22082
• 🎨 Sidebar user menu positioning. The sidebar user menu no longer drifts rightward when the sidebar is resized, keeping the menu properly aligned with its trigger. #​21853
• 💻 Code block UI. Code block headers are now sticky and properly positioned, with language labels now showing tooltips for truncated text. Commit
• 📊 Multi-model responses horizontal scroll. The model list in multi-model responses tabs now has horizontal scroll support, making all models accessible on desktop screens. #​21800, Commit
• 🎭 TailwindCSS gray color theme. Custom gray color palette is now correctly applied to the CSS root theme layer, fixing an issue where --color-gray-x variables were missing. #​21900, #​21899
• 📎 Broken documentation links. Fixed broken links in the backend config and admin settings that pointed to outdated documentation locations. #​21904
• 🔓 OAuth session token decryption. OAuth sessions are now properly detached from the database context before token decryption, preventing potential database session conflicts when reading encrypted tokens. #​21794
• 🕐 Chat timestamp i18n fix. Chat timestamps in the sidebar now display correctly, fixing an issue where the time ago format (e.g., "5m", "2h", "3d") was not being localized properly due to incorrect variable casing in the translation function. Commit
• 🍞 Model toast notification fix. Hiding or showing a single model now displays only one toast notification instead of two, removing the redundant generic "model updated" message when a specific action toast is shown. #​22079
• 📡 Offline mode embedding model fix. Open WebUI no longer attempts to download embedding models when in offline mode, fixing error logs that occurred when trying to fetch models that weren't cached locally. #​22106, #​21405

v0.8.5

Compare Source

[0.8.5] - 2026-02-23

Added

• ⌨️ Voice dictation shortcut. Users can now toggle voice dictation using Cmd+Shift+L (or Ctrl+Shift+L on Windows/Linux), making it faster to start and stop dictation without clicking the microphone button.

Fixed

• 🚫 Model access KeyError fix. The /api/models endpoint no longer crashes with a 500 error when models have incomplete info metadata missing the user_id field (e.g. models using global default metadata).
• 🔄 Frontend initialization resilience. The app layout now gracefully handles individual API failures during initialization (getModels, getBanners, getTools, getUserSettings, setToolServers) instead of blocking the entire page load when any single call fails.
• 🛡️ Backend config null safety. Language detection during app initialization no longer crashes when the backend config fetch fails, preventing a secondary cause of infinite loading.

v0.8.5

Compare Source

Added

• ⌨️ Voice dictation shortcut. Users can now toggle voice dictation using Cmd+Shift+L (or Ctrl+Shift+L on Windows/Linux), making it faster to start and stop dictation without clicking the microphone button.

Fixed

• 🚫 Model access KeyError fix. The /api/models endpoint no longer crashes with a 500 error when models have incomplete info metadata missing the user_id field (e.g. models using global default metadata).
• 🔄 Frontend initialization resilience. The app layout now gracefully handles individual API failures during initialization (getModels, getBanners, getTools, getUserSettings, setToolServers) instead of blocking the entire page load when any single call fails.
• 🛡️ Backend config null safety. Language detection during app initialization no longer crashes when the backend config fetch fails, preventing a secondary cause of infinite loading.

v0.8.4

Compare Source

[0.8.4] - 2026-02-23

Added

• 🛜 Provider URL suggestions. The connection form now displays a dropdown with suggested URLs for popular AI providers, making it easier to configure connections. Commit
• ☁️ Anthropic model fetching. The system now properly fetches available models from the Anthropic API, ensuring all Anthropic models are accessible. Commit
• 💡 No models prompt. When no models are available, a helpful prompt now guides users to manage their provider connections. Commit
• ⚙️ Connection enable/disable toggles. Individual provider connections can now be enabled or disabled from both admin and user settings. Commit
• ⏸️ Prompt enable/disable toggle. Users can now enable or disable prompts directly from the prompts list using a toggle switch, without needing to delete and recreate them. Inactive prompts display an "Inactive" badge and are still visible in the list. Commit
• 🗑️ Memory deletion. Agents can now delete specific memories that are no longer relevant, duplicated, or incorrect, giving better control over stored memory content. Commit
• 📋 Memory listing. Agents can now list all stored memories, enabling them to identify which memories to manage or delete based on the complete memory inventory. Commit
• 📦 Auto pip install toggle. Administrators can now disable automatic pip package installation from function frontmatter requirements using the ENABLE_PIP_INSTALL_FRONTMATTER_REQUIREMENTS environment variable, providing more control over function dependency management. Commit
• 🔗 Anthropic Messages API proxy. A new API endpoint now supports the Anthropic Messages API format, allowing tools like Claude Code to authenticate through Open WebUI and access configured models. Tool calls are now properly supported in streaming responses with correct multi-block indexing, and error status from tools is propagated correctly. The endpoint converts requests to OpenAI format internally, routes them through the existing chat pipeline, and returns responses in Anthropic format. #​21390, Commit, Commit
• 👥 Multi-device OAuth sessions. Users can now stay logged in on multiple devices simultaneously with OAuth, as re-logging in no longer terminates existing sessions. The oldest sessions are automatically pruned when the session limit is exceeded. #​21647, Commit
• 🔐 OAuth group default share setting. Administrators can now configure the default sharing setting for OAuth-created groups using the OAUTH_GROUP_DEFAULT_SHARE environment variable, allowing control over whether new groups default to private or shared with members. #​21679, Commit
• 🔧 Knowledge base import behavior. The web content import endpoint now supports a configurable overwrite flag, allowing users to add multiple URLs to the same knowledge base instead of replacing existing content. #​21613, #​21336, Commit
• 🧩 Skill JSON import support. Skills can now be imported from both JSON and Markdown files. #​21511
• 🔍 You.com web search provider. A new web search provider option for You.com is now available, giving users another search engine choice for web-enabled models. The You.com provider enriches search results by including both descriptions and snippets for better context. #​21599
• 🚀 Message list performance. Loading conversation history when sending messages is now significantly faster, improving response latency before the model starts generating. This also speeds up chat search and RAG context building. #​21588
• 🎯 Concurrent embedding request control. Administrators can now control the maximum number of concurrent embedding API requests using the RAG_EMBEDDING_CONCURRENT_REQUESTS environment variable, helping manage API rate limits while maintaining embedding performance. #​21662, Commit
• ⚡ Message upsert optimization. Loading chat data during message saving is now significantly faster by eliminating a redundant database call that occurred on every message upsert, which happens many times during streaming responses. [

---

Configuration

📅 Schedule: Branch creation - At any time (no schedule defined), Automerge - "after 2am,before 5am" (UTC).

🚦 Automerge: Disabled by config. Please merge this manually once you are satisfied.

♻ Rebasing: Whenever PR becomes conflicted, or you tick the rebase/retry checkbox.

🔕 Ignore: Close this PR and you won't be reminded about this update again.

---

• If you want to rebase/retry this PR, check this box

---

This PR was generated by Mend Renovate. View the repository job log.