mirror of
https://github.com/open-webui/open-webui.git
synced 2026-11-20 21:52:14 +03:00
docs: start 0.13.4 changelog (#33968)
* docs: start 0.13.4 changelog * docs: add Open folder button, agent titles, folder terminal, embedding folder, working folder and Docker lock entries to 0.13.4 changelog * docs: move working folder entry to Fixed and Docker lock entry to Added in 0.13.4 changelog * docs: add page tabs, per-chat terminal, agent sessions, shared chat, Tether upload and API entries to 0.13.4 changelog * docs: add self-service terminals, Terminals templates change, side chat reload fix and upload default to 0.13.4 changelog * docs: add faster folder reopening to 0.13.4 changelog * docs: update self-service terminals entry in 0.13.4 changelog * docs: add View terminal files and update tabs, terminals and chat terminal entries in 0.13.4 changelog * docs: add side chat, visualizations, mentions, sub-agent permission, agent saves, tab progress, grouped tools and menu entries to 0.13.4 changelog * docs: add visualization prompt setting and terminal picker for tabs to 0.13.4 changelog * docs: add side chat fixes, shared chat terminals, calendar and arena features, agent model IDs and fixes from merged PRs to 0.13.4 changelog * docs: add Git panel issues, tool calls in summaries and steadier Tether connections to 0.13.4 changelog * docs: extend Git panel entry in 0.13.4 changelog * docs: update 0.13.4 changelog date * docs: add Tether connections to 0.13.4 changelog * docs: add new fixes to 0.13.4 changelog * docs: add tab titles and account cleanup to 0.13.4 changelog * docs: add folder board and statuses to 0.13.4 changelog * docs: add agent reasoning effort to 0.13.4 changelog * docs: add folder automations to 0.13.4 changelog * docs: drop folder terminal scope from 0.13.4 changelog * docs: add automation page editing to 0.13.4 changelog * docs: add folder mark as read fix to 0.13.4 changelog * docs: add live Files tab to 0.13.4 changelog * docs: update 0.13.4 changelog date * docs: extend tag entry in 0.13.4 changelog * docs: update 0.13.4 changelog date * docs: add live permission fix to 0.13.4 changelog * docs: add Tether MCP servers and more live permission fixes to 0.13.4 changelog * docs: add coding agent command and edit views to 0.13.4 changelog * docs: add new agents, fixes and perf to 0.13.4 changelog * docs: extend coding agent steps and overview tab in 0.13.4 changelog * docs: add side chat menu, terminal steps and new fixes to 0.13.4 changelog * docs: link fixed issues in 0.13.4 changelog * docs: add terminal instance view and temporary chat fix to 0.13.4 changelog * docs: add folder path typing, agent session note and fixes to 0.13.4 changelog * docs: changelog update for markdown outline, links in messages, folder drafts and statuses * docs: changelog update for files menu placement and MCP OAuth headers * docs: changelog update for folder knowledge, skill file drops, Firefox scrolling, phone keyboard and sub-agent reports * docs: changelog update for queued messages, agent regeneration, failed tool labels, context lists, note recordings and toolbar changes * docs: changelog update for default terminal, agent forks, compaction, parameters, MCP reconnects, sheets, SSL checks and notifications * docs: changelog update for looping chats, tool timeouts, agent images and large MCP tool lists * docs: changelog update for terminal lock during replies and file cards following the terminal * docs: drop emojis from the 0.13.4 changelog * docs: changelog update for terminal skill mentions and OpenAI image defaults * docs: changelog date * docs: changelog update for the Android comment box and model selection without Multiple Models * docs: changelog update for OpenAI-compatible vector stores and optional valves * docs: list vector stores as knowledge bases * docs: add image connections * docs: add video generation and OAuth session fixes * docs: move UI rearrangements out of Changed * docs: add sign-in, Markdown editor and settings fixes * docs: add MCP sign-in, group name and model list fixes
This commit is contained in:
committed by
GitHub
Unverified
parent
f2a30994ca
commit
7e3ecfa59b
+158
@@ -5,6 +5,164 @@ All notable changes to this project will be documented in this file.
|
||||
The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.1.0/),
|
||||
and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
|
||||
|
||||
## [0.13.4] - 2026-11-17
|
||||
|
||||
### Added
|
||||
|
||||
- **Pages open as tabs.** Notes, channels, the workspace, automations, the calendar, the playground, the home page and new chats from pinned models now open as tabs beside the current chat when you click or drag them from the sidebar, pick them in the user menu or follow a link to one of them in a chat message, and starting a new chat, also from a sidebar folder, keeps the open tabs.
|
||||
- **Full chat in side chats.** Side chats now run the same chat as the main view, so message buttons, sources, token usage, follow-up suggestions, slash commands, queued messages, chat variables, web search settings, tool events such as code execution and input requests, HTML and SVG previews, notifications while the window is in the background, unsent text after a page reload and answers from several models at once now work there too, and a side chat's tab now has the chat menu. [#34064](https://github.com/open-webui/open-webui/pull/34064), [#34165](https://github.com/open-webui/open-webui/pull/34165), [#34147](https://github.com/open-webui/open-webui/pull/34147), [#34031](https://github.com/open-webui/open-webui/issues/34031)
|
||||
- **Queued messages steer replies.** A message you send to your own chat while a reply is still running now waits on the server, so it is kept when you leave the page, and the reply wraps up at its next step so your message is answered right away with what it has done so far.
|
||||
- **Interactive visualizations.** Models can now draw interactive charts, diagrams, calculators and explainers right in the chat with the new Visualizations built-in tool, which is off by default and turned on per model under Builtin Tools, and administrators can rewrite its instructions under Admin Settings > Interface > Visualization Prompt or with "VISUALIZATION_PROMPT_TEMPLATE".
|
||||
- **Folder chat board.** Folder pages now list their chats with a search through titles and messages and filters for status, tags, people, activity and last update, and can show them as a board with columns and rows by status, tag or activity such as unread, generating or waiting for approval, gives each chat there the sidebar's chat menu to rename, move, archive or delete it, and a folder's badge in the sidebar now also counts chats waiting for approval. [#33771](https://github.com/open-webui/open-webui/issues/33771)
|
||||
- **Chat statuses in folders.** Folders can now set up statuses for their chats under Edit statuses, with Backlog, To do, In progress, Waiting and Done offered as suggestions, chats then get a status from their menu on the folder page or by dragging them between board columns, people who can edit a shared folder can set the status of every chat in it, and each chat shows how long it has been in its status.
|
||||
- **Colored chat tags.** Tags can now get a color on the folder page that applies wherever the tag is used, and the chat menu has a Tags submenu to find, add and create tags, and people who can edit a shared folder can tag every chat in it.
|
||||
- **Folder automations.** Folder pages now have an Automations section that lists the folder's automations with their state and last run and creates new ones there, chats made by an automation are marked with a clock, and automations can now post into shared folders their creator can edit, where everyone who can edit the folder can pause and run them and see their runs.
|
||||
- **Knowledge on folder pages.** Folder pages now have a Knowledge section that lists the knowledge bases, notes, chats and files new chats in the folder can use, where you can add, upload, drop, open and remove them.
|
||||
- **Edit automations in place.** An automation's page now edits its name, instructions, status, schedule, model, terminal and destination right there with a Save button that appears once something changed, shows saved instructions collapsed behind Show more, lists its runs as activity below, and has Copy link and Duplicate in its More menu.
|
||||
- **Self-service terminals.** With a Terminals server connected, users who have Write access to the connection can now create terminals under Terminals in the terminal menu of the message input, which shows for them even before they have a terminal, choose whether everyone shares one instance or each user, chat or model gets its own, set a start directory, a setup script whose progress and failures show in System info, the image, CPU, memory, storage, network profile and environment variables, share a terminal with users and groups, and start, stop, restart or delete its instances, and those who manage a terminal can also open the system info and files of other people's running instances.
|
||||
- **Your own connections through Tether.** Models of OpenAI-compatible connections, OpenAPI and MCP tool servers and terminals that you add in the Tether app on your computer now show up for you in the model selector, the Tools list of the + menu and the terminal menu, with Open WebUI reaching them through Tether so their addresses, commands and keys stay on your computer. [#34397](https://github.com/open-webui/open-webui/issues/34397)
|
||||
- **Chats remember their terminal.** Each chat now saves its terminal and working folder and switches back to them when you reopen it, including a terminal you turned off, while chats from earlier versions open without a terminal unless they are coding agent chats.
|
||||
- **Default terminal for all models.** Administrators can now pick a default terminal under Admin Settings > Models > Model Defaults, which new chats start with for every model that has no terminal of its own.
|
||||
- **Shared chat terminals.** People who can reply to a chat shared with Allow replies can now use that chat's own terminal in their messages, as the share dialog now points out. [#34103](https://github.com/open-webui/open-webui/issues/34103)
|
||||
- **Git panel issues and actions.** When the terminal supports it, the Git panel now also lists the repository's issues to browse, create, edit, comment on, close and reopen, the pull request view adds a status filter, activity, conflict status and a choice of merge method, searching issues and pull requests now searches the whole repository instead of only the loaded list, changed files can be picked with Ctrl or Shift click to discard several at once or all of them, and buttons such as Work on this, Fix, Address feedback and Resolve conflicts in chat write a ready-made request for the model into the message input.
|
||||
- **Several image models.** Administrators can now add several image connections under Admin Settings > Images, each an OpenAI-compatible server such as LiteLLM, Gemini, ComfyUI or AUTOMATIC1111 with its own models, sizes, workflows and access, used for generating images, editing them or both, where OpenAI-compatible ones offer the same sign-in options and custom headers as model connections and list the image models of a LiteLLM server by themselves when no model IDs are entered, choose the default generation and edit models and give a model its own default image model in its settings, and users with more than one image model pick one from the Image entry in the + menu of the message input. [#10344](https://github.com/open-webui/open-webui/discussions/10344)
|
||||
- **Video generation.** Models can now generate videos with a built-in tool, also starting from an image in the chat, once an administrator turns on Video Generation under Admin Settings > Videos, adds connections to OpenAI, LiteLLM or OpenRouter, which offers models such as Veo and Kling, with a default video model, per-model defaults and access settings, and grants the Video Generation permission, which is off by default, after which users turn on Video in the + menu, can pick the model there and watch the result in the chat. [#13780](https://github.com/open-webui/open-webui/discussions/13780), [#13927](https://github.com/open-webui/open-webui/discussions/13927)
|
||||
- **Models can mention people.** With the Chat History built-in tool, models can now look up users by name or email and mention them in a reply, and each mentioned user who can see the chat gets an inbox notification that opens the chat at that reply.
|
||||
- **Sub-agent permission.** Administrators can now limit sub-agents to chosen users and groups with a Sub-agents switch under Features in the default and group permissions, or with "USER_PERMISSIONS_FEATURES_SUBAGENTS", and users without it get neither the sub-agent nor the timer tool. [#34040](https://github.com/open-webui/open-webui/pull/34040), [#34038](https://github.com/open-webui/open-webui/issues/34038)
|
||||
- **Smoother replies with code blocks.** Long replies with many code blocks no longer stall the tab a little longer with every new code block while they stream, so a reply with 300 code blocks that arrives at once now shows in 9 seconds instead of more than two minutes. [#34495](https://github.com/open-webui/open-webui/pull/34495), [#34492](https://github.com/open-webui/open-webui/issues/34492)
|
||||
- **Faster folder reopening.** Reopening a folder now shows its chats, details and coding agent models right away from what was already loaded and refreshes them in the background, sidebar folders keep their loaded chats when collapsed and expanded again, and folder chat lists update immediately when a chat is renamed, moved, archived or deleted.
|
||||
- **Faster model list with Redis.** With Redis, the model list now loads faster and Redis no longer keeps a copy of every model profile image, which helps most with large animated or imported images, and Functions and Tools now receive the model without its profile image. [#34595](https://github.com/open-webui/open-webui/pull/34595), [#34594](https://github.com/open-webui/open-webui/issues/34594)
|
||||
- **Readable agent and terminal steps.** Steps of coding agents and the terminal tool calls of other models now show in the chat as what they did, a command with its output and its exit code when it failed, a diff of each edited file with its added and removed lines, or the files read, code and web searches, folders listed and pages fetched, and a group of steps is summarized as for example Ran 2 commands, Edited 1 file, Read 3 files.
|
||||
- **Lighter coding agent replies.** While a coding agent writes its reply, the chat is no longer saved in full about once a second, so long agent chats write several times less to the database for each reply. [#34062](https://github.com/open-webui/open-webui/issues/34062)
|
||||
- **Reasoning effort for coding agents.** The reasoning effort set in the advanced parameters now also applies to Claude Code, Pi and the coding agents connected over ACP such as Cursor and Gemini, which stop with an error when the chosen level is not offered for the model. [#34167](https://github.com/open-webui/open-webui/issues/34167)
|
||||
- **More coding agents.** Hermes, Antigravity, GitHub Copilot, Goose and Qwen Code now work as coding agents, as do other ACP agents added to Open Terminal with "OPEN_TERMINAL_ACP_AGENTS", and choosing Auto for tool permissions now lets Codex, Claude Code and Grok approve routine actions themselves and pass on only what they escalate, with Open Terminal 0.17.19 or later for the new agents and for Claude Code and Grok.
|
||||
- **Regenerate in coding agent chats.** Replies in coding agent chats can now be regenerated and earlier messages edited, and a forked or cloned coding agent chat picks up where the original left off, with Codex and Claude Code continuing their own conversation from that point on an up-to-date Open Terminal and other agents starting over from the chat so far, while file changes the agent already made stay.
|
||||
- **Instant first message.** The first message of a new chat now shows up right away while the chat is still being created, instead of only once it exists, unless files are still being uploaded with it.
|
||||
- **Reply progress on tabs.** The chat tab and side chat tabs now show a spinner while a reply is being written, so you can see from another tab when an answer is still coming.
|
||||
- **Rename from the tab.** Double-clicking the chat tab now renames the chat, also a chat someone else made in a folder you own, and on a folder page it renames the folder.
|
||||
- **Chat menu items moved.** Overview, Artifacts and Comments are no longer in the chat's menu and now open from the + menu of the tab bar, where Overview opens as a tab beside the chat instead of a full-screen window.
|
||||
- **Share folders from their menu.** The folder menu in the sidebar and on the folder page now has a Share option for administrators and for owners allowed to share folders.
|
||||
- **View terminal files.** Terminals in the terminal menu of the message input now have a More menu with System info and a new View files option, which opens that terminal's files in a dialog without selecting the terminal for the chat.
|
||||
- **Live Files tab.** With a terminal that reports file changes, the Files tab now updates by itself without a loading spinner when files in the folders it shows change, also through commands run outside the chat.
|
||||
- **Type a folder path.** Clicking the path above a Files tab now lets you type a folder to go to, absolute or relative to the current one, with Tab completing folder names, and with a file open it shows the file's full path to select and copy.
|
||||
- **Markdown outline in Files.** Markdown files shown in a terminal's Files preview now have an Outline button that lists their headings and jumps to the one you pick. [#34494](https://github.com/open-webui/open-webui/pull/34494), [#34525](https://github.com/open-webui/open-webui/discussions/34525)
|
||||
- **Terminal picker for tabs.** When no terminal is selected for the chat, the menu for opening a new tab now still offers Files, new files, terminals, the browser, the desktop and voice notes, and picking one first asks you to choose a terminal from a searchable list.
|
||||
- **Terminal tab titles.** Terminal tabs are now named after the title that the program running in them sets, or after the running program when the terminal reports it, and show an icon for well-known programs such as Claude Code, Codex, Python or Node.js.
|
||||
- **Grouped tools and skills.** The Tools and Skills lists in the + menu of the message input are now grouped by where each entry comes from, such as the workspace, tool servers or the terminal. [#34085](https://github.com/open-webui/open-webui/pull/34085)
|
||||
- **Labels on failed tool calls.** Tool calls that failed now keep the short label the model gave them, also in a group of calls, where one failed call used to hide every label. [#34000](https://github.com/open-webui/open-webui/pull/34000), [#34529](https://github.com/open-webui/open-webui/discussions/34529)
|
||||
- **Drop files into skills.** Files and folders dragged onto a skill's files in the workspace now upload into the open folder, like the upload buttons do.
|
||||
- **Editable calendars.** Calendars can now be renamed and given another color after they are created, from a pencil next to each calendar in the Calendar sidebar. [#34136](https://github.com/open-webui/open-webui/pull/34136)
|
||||
- **Note comments button.** Comments on a note now open from a button in the note's top bar instead of from its menu, and in a note you can edit, commenting on selected text is now a Comment button in the formatting toolbar.
|
||||
- **One heading menu.** The formatting toolbar of the message input and of notes now offers Heading 1, 2 and 3 in a single heading menu instead of three separate buttons.
|
||||
- **Arena models view.** The view filter in Admin Settings > Models now has an Arena Models option that lists only the arena models. [#34137](https://github.com/open-webui/open-webui/pull/34137)
|
||||
- **Clone arena models.** Each arena model in Admin Settings > Evaluations now has a Clone button that opens a new arena model filled in from it. [#34138](https://github.com/open-webui/open-webui/pull/34138)
|
||||
- **Model names in analytics.** Admin analytics now name coding agent models by agent and model for replies made from now on, and the tooltips of the session timeline show model names instead of model IDs.
|
||||
- **OpenAI-compatible vector stores.** External knowledge sources under Admin Settings > Integrations can now be OpenAI-compatible vector stores on OpenAI, Azure OpenAI or LiteLLM, searched on the server so no matching embedding model is needed, and you either pick one or more stores for a knowledge base or turn on 'List stores as knowledge bases' to show every store of the connection, or only chosen ones, as read-only knowledge bases that stay in sync with the endpoint and share the connection's access settings.
|
||||
- **Open folder button.** Folders in the working directory menu of the message input now have an Open folder button that opens the folder in a new Files tab.
|
||||
- **Locked Docker dependencies.** The standard and slim Docker images now install every Python package except PyTorch from lock files pinned with hashes, so rebuilding the same version installs the same packages, and custom images that edit "backend/requirements.txt" must regenerate the locks with "backend/update-locks.sh" for the edit to take effect. [#33981](https://github.com/open-webui/open-webui/pull/33981), [#33979](https://github.com/open-webui/open-webui/issues/33979)
|
||||
- **OAuth session removals logged.** Removing a stored OAuth session because of the "OAUTH_MAX_SESSIONS_PER_USER" limit, a sign-out sent by the identity provider or a reconnected MCP server now logs a warning naming the user, the session and the reason. [#34583](https://github.com/open-webui/open-webui/pull/34583), [#34580](https://github.com/open-webui/open-webui/issues/34580)
|
||||
- **Accessibility improvements.** A tab moved with Alt+Shift+Arrow keys now keeps the keyboard focus, so it can be moved several places in a row.
|
||||
- **General improvements.** Various improvements were implemented across the application to enhance performance, stability, and security.
|
||||
- **Translation updates.** Translations for Azerbaijani, Basque, Bengali, Bosnian, Bulgarian, Cebuano, Croatian, Czech, Danish, Dutch, Estonian, Faroese, Filipino, Galician, Georgian, German, Greek, Hebrew, Hindi, Hungarian, Indonesian, Irish, Italian, Japanese, Kabyle, Korean, Latvian, Lithuanian, Malay, Norwegian Bokmål, Persian, Polish, Portuguese (Brazil), Punjabi, Romanian, Russian, Serbian, Slovak, Slovenian, Spanish (Spain), Swedish, Tamil, Thai, Tibetan, Turkish, Turkmen, Ukrainian, Urdu, Uyghur and Vietnamese were enhanced and expanded.
|
||||
|
||||
### Fixed
|
||||
|
||||
- **Temporary chats leave no memories.** The background memory review now skips temporary chats, so facts from them are no longer saved as memories that show up in later chats. [#34263](https://github.com/open-webui/open-webui/pull/34263), [#34262](https://github.com/open-webui/open-webui/issues/34262)
|
||||
- **Temporary chat choice kept.** With Temporary Chat by default turned on in the interface settings, a new chat switched to a normal chat now stays normal after a page reload.
|
||||
- **Chat keeps the newest turn.** A window that missed a turn, such as a laptop that just woke up, no longer moves the chat back to an older turn when it saves an edited message, so the newest reply stays in the conversation. [#34160](https://github.com/open-webui/open-webui/pull/34160), [#34126](https://github.com/open-webui/open-webui/issues/34126)
|
||||
- **Smooth Firefox scrolling.** Scrolling up through a long chat in Firefox no longer throws the view back down, sometimes by several screens, while older messages load. [#33998](https://github.com/open-webui/open-webui/pull/33998), [#33994](https://github.com/open-webui/open-webui/issues/33994)
|
||||
- **No keyboard pop-up on phones.** On a phone, opening a chat, switching chats or starting a new chat no longer puts the cursor in the message input and brings up the on-screen keyboard over the conversation. [#34059](https://github.com/open-webui/open-webui/issues/34059)
|
||||
- **Chats with looping history.** A chat whose saved messages loop back on themselves now opens and shows its messages instead of freezing the browser tab until it crashes. [#34481](https://github.com/open-webui/open-webui/pull/34481), [#34477](https://github.com/open-webui/open-webui/issues/34477)
|
||||
- **Summary kept through compaction.** A conversation compacted again between tool calls now keeps what its earlier summary held, and a chat compacted after such a compaction goes on from its newer summary instead of falling back to the older one and sending more than the model can take. [#34151](https://github.com/open-webui/open-webui/pull/34151), [#34141](https://github.com/open-webui/open-webui/issues/34141), [#34140](https://github.com/open-webui/open-webui/issues/34140)
|
||||
- **Tool calls in summaries.** Conversation summaries written with the plain summary prompt, as for models with a filter, now include the tool calls of earlier replies in saved chats. [#34139](https://github.com/open-webui/open-webui/issues/34139)
|
||||
- **Steadier context compaction.** Compacting a chat by hand right after a failed reply now keeps the summary for the next message instead of losing it along with the failed reply, and automatic compaction now also counts a tool result that ended a reply, such as one too long for the model to answer, so the next message compacts the chat instead of sending it in full. [#34434](https://github.com/open-webui/open-webui/issues/34434)
|
||||
- **Fewer repeated context lists.** With native function calling, the list of available skills and the attached knowledge are no longer added to the chat again before every reply, only when they changed or after compaction, so long chats use fewer tokens and need compaction later. [#34257](https://github.com/open-webui/open-webui/issues/34257)
|
||||
- **Final answers on vLLM.** When tool use stops, the request for a final answer no longer ends in a 400 error on vLLM, and Ollama models are kept from calling tools in it. [#34176](https://github.com/open-webui/open-webui/pull/34176), [#34169](https://github.com/open-webui/open-webui/issues/34169)
|
||||
- **Tool call turns without text.** Replies in which a model called a tool without writing any text are now sent back to the model without an empty text, so proxies such as LiteLLM no longer turn them into a placeholder line that the model then starts copying into its answers. [#34265](https://github.com/open-webui/open-webui/pull/34265), [#34264](https://github.com/open-webui/open-webui/issues/34264)
|
||||
- **Timed-out tool calls explained.** A terminal command or tool server call that runs past "AIOHTTP_CLIENT_TIMEOUT_TOOL_SERVER" now reaches the model as a timeout error instead of an empty one, so the model can tell you the call timed out. [#34119](https://github.com/open-webui/open-webui/pull/34119), [#34096](https://github.com/open-webui/open-webui/issues/34096)
|
||||
- **Sub-agent reports stay together.** Two background sub-agent reports that finish close together no longer split the chat into two branches that hide the first report and the reply to it. [#34542](https://github.com/open-webui/open-webui/pull/34542), [#34540](https://github.com/open-webui/open-webui/issues/34540)
|
||||
- **Sub-agents with browser questions.** A sub-agent no longer hangs on "Executing Sub-agent" when a filter or tool asks the browser something, because the script, confirmation or form now appears in the tab that has the chat open. [#34045](https://github.com/open-webui/open-webui/pull/34045), [#34041](https://github.com/open-webui/open-webui/issues/34041)
|
||||
- **Ollama proxy tool calls.** OpenAI-compatible clients that send back an assistant turn with only tool calls and no text no longer get a 400 "Field required" error from the Ollama OpenAI-compatible chat endpoint, so multi-turn tool calling works through it. [#34442](https://github.com/open-webui/open-webui/pull/34442), [#34441](https://github.com/open-webui/open-webui/issues/34441)
|
||||
- **Whole numbers in parameters.** Whole-number fields in the advanced parameters, such as Context Length, Max Tokens and Seed, now reject a value with a decimal point, such as "32.767" typed for 32767, instead of saving it and running the model with a context of only a few tokens. [#34416](https://github.com/open-webui/open-webui/issues/34416)
|
||||
- **Optional valves set to Custom.** Switching an optional valve, such as an on/off setting, from Default to Custom in the Valves section of the chat controls no longer fails with "Error updating function user valves", and emptying an optional on/off or number valve sets it back to its default. [#34478](https://github.com/open-webui/open-webui/pull/34478), [#34475](https://github.com/open-webui/open-webui/issues/34475)
|
||||
- **Math in table headers.** Formulas in table headers now keep their letters as written instead of being capitalized along with the header text. [#34245](https://github.com/open-webui/open-webui/pull/34245), [#34238](https://github.com/open-webui/open-webui/issues/34238)
|
||||
- **Math next to hyphens.** Inline formulas written directly next to a hyphen or non-breaking hyphen, as some models write them, now render instead of staying plain text or showing a math error. [#34244](https://github.com/open-webui/open-webui/pull/34244), [#34237](https://github.com/open-webui/open-webui/issues/34237)
|
||||
- **Coding agent chat titles.** Chats with a coding agent now get a generated title once the agent has replied, instead of keeping the default title. [#33971](https://github.com/open-webui/open-webui/pull/33971), [#33970](https://github.com/open-webui/open-webui/issues/33970)
|
||||
- **Agent model switches keep sessions.** Switching a coding agent chat to another model of the same agent now resumes the agent's existing session and working folder instead of starting a new session.
|
||||
- **Readable coding agent errors.** When the terminal refuses to start a coding agent, the reply now shows a short error instead of a technical dump containing the whole system prompt. [#34173](https://github.com/open-webui/open-webui/pull/34173), [#34172](https://github.com/open-webui/open-webui/issues/34172)
|
||||
- **Images reach coding agents.** Images attached to a message in a coding agent chat now reach the agent with the message instead of being dropped without a word. [#33993](https://github.com/open-webui/open-webui/pull/33993), [#33992](https://github.com/open-webui/open-webui/issues/33992), [#33991](https://github.com/open-webui/open-webui/discussions/33991)
|
||||
- **Terminal choice kept in folders.** In a new chat inside a folder with a default terminal, a terminal you pick or turn off now stays that way when the model changes, instead of going back to the folder default, and a new chat in a folder without a default terminal keeps the terminal you had selected instead of turning it off. [#33973](https://github.com/open-webui/open-webui/issues/33973)
|
||||
- **Terminal locked during replies.** The terminal and working folder pickers in the message input now stay locked while a reply is still being written, so the chat's terminal can no longer be switched in the middle of a reply. [#34383](https://github.com/open-webui/open-webui/issues/34383)
|
||||
- **File cards follow the terminal.** A file shown inline in a reply now gets its preview, Download and Open back as soon as its terminal is picked again, and stops reaching the file once another terminal is picked, instead of keeping its first state until a page reload. [#34101](https://github.com/open-webui/open-webui/issues/34101)
|
||||
- **Terminal skill mentions after reload.** A sent message that mentions a skill of the picked terminal now still shows it as a chip after a page reload, and the + menu lists that terminal's skills right away, instead of both only working after the skill list was opened once. [#34102](https://github.com/open-webui/open-webui/issues/34102)
|
||||
- **Working folder stays put.** Browsing to another folder in a Files tab no longer changes the working folder shown in the working directory menu and used by a new chat, and opening Files again shows an existing Files tab where you left it instead of moving it to the working folder.
|
||||
- **Unsaved terminal file edits kept.** An edit not yet saved in a file opened from a terminal's Files tab now comes back with its unsaved mark after you open another chat and return, and a workbook saved while you leave the chat keeps the worksheet you had open. [#34135](https://github.com/open-webui/open-webui/pull/34135), [#34107](https://github.com/open-webui/open-webui/issues/34107)
|
||||
- **File sizes after Refresh.** In a terminal's file viewer, Refresh and the back and forward arrows now keep the open file's real size instead of showing 0 B, and Refresh shows the new size when the file changed. [#34459](https://github.com/open-webui/open-webui/pull/34459), [#34458](https://github.com/open-webui/open-webui/issues/34458)
|
||||
- **Files menu near edges.** The menu from a right click on the empty part of a Files list now opens upwards or to the left when there is no room below or beside the pointer, so all of its entries stay inside the window. [#34125](https://github.com/open-webui/open-webui/pull/34125), [#34097](https://github.com/open-webui/open-webui/issues/34097)
|
||||
- **Long Markdown files.** Opening a long Markdown file from a terminal's Files, or typing in one of its code blocks, no longer freezes the page for many seconds, because code blocks without a known language now show as plain text instead of being checked against every language, which also keeps notes and the rich text message input responsive. [#34491](https://github.com/open-webui/open-webui/pull/34491), [#34486](https://github.com/open-webui/open-webui/issues/34486)
|
||||
- **Shared chat changes saved.** Someone replying in a chat shared with Allow replies can now change its model, terminal and working folder, and the choice is saved to the chat instead of being dropped when it is reopened.
|
||||
- **Permission changes reach open pages.** When an admin changes permissions, open pages now follow without a reload, so a new chat that is open becomes a temporary chat once Enforce Temporary Chat is turned on instead of failing on the first message, Compare in the model selector turns off and only the first selected model stays once Multiple Models in Chat is turned off, an open channel shows the permission notice in place of the channel once Channels is turned off, and turning off Folders removes Move from the chat menu and closes an open Create Folder window. [#34513](https://github.com/open-webui/open-webui/pull/34513), [#34512](https://github.com/open-webui/open-webui/issues/34512), [#34515](https://github.com/open-webui/open-webui/pull/34515), [#34514](https://github.com/open-webui/open-webui/issues/34514), [#34511](https://github.com/open-webui/open-webui/pull/34511), [#34510](https://github.com/open-webui/open-webui/issues/34510), [#34509](https://github.com/open-webui/open-webui/pull/34509), [#34508](https://github.com/open-webui/open-webui/issues/34508), [#34507](https://github.com/open-webui/open-webui/pull/34507), [#34506](https://github.com/open-webui/open-webui/issues/34506)
|
||||
- **Side chats after reload.** A saved side chat tab restored after a page reload or on another device now shows only the side conversation again, instead of also listing the messages copied from the parent chat. [#34002](https://github.com/open-webui/open-webui/issues/34002)
|
||||
- **Your devices list updates live.** A computer that connects through Tether after the page has loaded now appears under Your devices in the terminal menu right away, and the menu also picks up disconnects and device name changes, instead of needing a page reload.
|
||||
- **Steadier Tether connections.** Requests to a computer connected through Tether are no longer cut off after five minutes while data keeps flowing, read-only requests interrupted by a brief disconnect are repeated once Tether reconnects, and a busy computer now refuses new operations right away instead of letting some of them time out.
|
||||
- **Smooth Browser tab resizing.** Dragging the divider of the side panel no longer makes the Browser tab flicker and resize the page in slow motion. [#33989](https://github.com/open-webui/open-webui/pull/33989), [#33988](https://github.com/open-webui/open-webui/issues/33988)
|
||||
- **Browser tab in two windows.** With the same chat open in two windows, a Browser tab in Chrome mode no longer keeps reconnecting in both, and the window that lost the tab shows that the browser opened in another viewer. [#34133](https://github.com/open-webui/open-webui/pull/34133), [#34095](https://github.com/open-webui/open-webui/issues/34095)
|
||||
- **Ports open in new tabs.** The Open in new tab button beside a port on the start page of a Browser tab now opens the server running on that port instead of a "404: Not Found" page. [#34120](https://github.com/open-webui/open-webui/pull/34120), [#34108](https://github.com/open-webui/open-webui/issues/34108)
|
||||
- **Steadier menu placement.** Menus that open above their button, such as Model controls and the + menu of the message input, now grow upward when their content changes instead of briefly covering the message input, and right-aligned menus in Firefox no longer creep sideways while they open. [#34052](https://github.com/open-webui/open-webui/pull/34052), [#34051](https://github.com/open-webui/open-webui/issues/34051), [#34227](https://github.com/open-webui/open-webui/pull/34227), [#34226](https://github.com/open-webui/open-webui/issues/34226)
|
||||
- **Folder page kept.** Opening a chat in another browser tab no longer sends a folder's chat list back to page 1. [#34145](https://github.com/open-webui/open-webui/pull/34145), [#34142](https://github.com/open-webui/open-webui/issues/34142)
|
||||
- **Drafts kept per folder.** Unsent text in a new chat now stays with the folder or home page you typed it on, so opening another folder shows that folder's own unsent text instead of bringing yours along.
|
||||
- **Mark folder chats read.** Mark all as read in the menu on a folder page now marks the folder's chats as read instead of doing nothing. [#34404](https://github.com/open-webui/open-webui/pull/34404), [#34403](https://github.com/open-webui/open-webui/issues/34403)
|
||||
- **Same-folder chat drops.** Dropping a chat back on the folder it is already in no longer moves it there again and reloads every folder in the sidebar. [#34461](https://github.com/open-webui/open-webui/pull/34461), [#34460](https://github.com/open-webui/open-webui/issues/34460)
|
||||
- **Links in messages.** A link to a folder in a chat or channel message now opens the folder in Open WebUI instead of a new browser tab, and Ctrl or Cmd clicking a link to a chat, note or channel in a message now opens it in a new browser tab instead of the current one. [#34464](https://github.com/open-webui/open-webui/issues/34464)
|
||||
- **Tagging chats in parallel.** Adding the same new tag to several chats at the same time, as an API client tagging a batch of chats or two chat imports can do, no longer fails with a server error for all but one chat and leaves the others untagged. [#34144](https://github.com/open-webui/open-webui/pull/34144), [#34143](https://github.com/open-webui/open-webui/issues/34143)
|
||||
- **Hover preview behind menus.** A chat's hover preview in the sidebar no longer opens on top of another chat's open menu and covers its items. [#34231](https://github.com/open-webui/open-webui/pull/34231), [#34230](https://github.com/open-webui/open-webui/issues/34230)
|
||||
- **Hover preview without bubbles.** With Chat Bubble UI turned off, your messages in a chat's hover preview now show as padded bubbles instead of text squeezed onto a grey block. [#34347](https://github.com/open-webui/open-webui/issues/34347)
|
||||
- **Thread reply notifications.** Clicking the browser notification for a reply in a channel thread now opens that thread instead of the channel with the thread closed. [#34233](https://github.com/open-webui/open-webui/pull/34233), [#34232](https://github.com/open-webui/open-webui/issues/34232)
|
||||
- **Steady Tools count.** The count on the Tools tab of the workspace no longer drops to 0 for a moment when the Tools page opens. [#34236](https://github.com/open-webui/open-webui/pull/34236)
|
||||
- **Knowledge base labels.** The knowledge picker in the chat, the knowledge selector of the model editor and attached knowledge bases now call them Knowledge or Knowledge Base instead of Collection, the vector database term, and the search box inside a knowledge base now says Search Files. [#34432](https://github.com/open-webui/open-webui/pull/34432)
|
||||
- **Image zoom reset.** Resetting the zoom of an image preview from below 10% now returns it to 100% and centers it again. [#34080](https://github.com/open-webui/open-webui/pull/34080), [#34079](https://github.com/open-webui/open-webui/issues/34079)
|
||||
- **OpenAI image generation defaults.** Generating images with OpenAI now works with the stock settings, because GPT image models get the auto size instead of the old 512x512 default they reject and Auto can be saved with the model field left empty, and a failed request now shows the provider's own error, such as "Invalid size".
|
||||
- **Notes with footnotes.** Notes containing footnote references such as "[^1]" now open and show their content instead of failing with an error. [#33987](https://github.com/open-webui/open-webui/pull/33987), [#33986](https://github.com/open-webui/open-webui/issues/33986)
|
||||
- **Note recordings on S3 storage.** Recordings made in a note now upload to S3-compatible storage that rejects colons in file names and are named like "Recording-2026-11-16 15-25-06.webm" in every language, instead of after a date that slashes could cut short. [#34545](https://github.com/open-webui/open-webui/pull/34545), [#34543](https://github.com/open-webui/open-webui/issues/34543)
|
||||
- **Comment box on Android.** Writing a comment on selected text on an Android phone no longer closes the comment box as soon as the keyboard opens.
|
||||
- **Unpin tooltip on notes.** The Unpin button on a pinned note in the sidebar now shows a tooltip, as it does on pinned models. [#34568](https://github.com/open-webui/open-webui/pull/34568), [#34567](https://github.com/open-webui/open-webui/issues/34567)
|
||||
- **Translated automation schedules.** The Schedule button in the dialog that creates an automation now shows the schedule in the chosen language instead of always in English. [#34428](https://github.com/open-webui/open-webui/pull/34428), [#34426](https://github.com/open-webui/open-webui/issues/34426)
|
||||
- **PowerPoint decks with sections.** PowerPoint files that use sections now preview in the chat and the knowledge base file viewer and show their comments, instead of failing with "Missing slide relationship". [#34276](https://github.com/open-webui/open-webui/pull/34276), [#34270](https://github.com/open-webui/open-webui/issues/34270)
|
||||
- **Large workbook sheets.** Excel workbooks with a sheet too large to preview now still open with their sheet tabs in a terminal's Files tab, the chat and the knowledge base, where that sheet says it is too large to preview and can be downloaded, instead of the whole file failing to preview. [#34247](https://github.com/open-webui/open-webui/issues/34247)
|
||||
- **Local embedding model folders.** With the embedding model set to a folder on disk, saving Admin Settings > Documents no longer shows a "Model filesystem path detected" error and now also saves the embedding settings such as the batch size, which were skipped before. [#33978](https://github.com/open-webui/open-webui/pull/33978), [#33977](https://github.com/open-webui/open-webui/issues/33977)
|
||||
- **Fuller web page text.** With the Playwright web loader, the text of definition lists, of figures with their captions and of collapsible sections, open or closed, now reaches the model, so attached web pages and web search results no longer lose content such as the function and parameter descriptions on many documentation sites. [#34282](https://github.com/open-webui/open-webui/pull/34282), [#34281](https://github.com/open-webui/open-webui/issues/34281), [#34277](https://github.com/open-webui/open-webui/pull/34277), [#34272](https://github.com/open-webui/open-webui/issues/34272)
|
||||
- **Unreachable sites skipped sooner.** With SSL verification on, the Playwright, Tavily and Microsoft Web IQ web loaders now skip a site that never answers after 10 seconds instead of waiting about two minutes per link. [#34490](https://github.com/open-webui/open-webui/pull/34490), [#14241](https://github.com/open-webui/open-webui/discussions/14241)
|
||||
- **Web pages behind internal certificates.** With SSL verification on, the Playwright, Tavily and Microsoft Web IQ web loaders now load pages signed by an internal certificate authority set through "AIOHTTP_CLIENT_SSL_CERT_FILE" or "AIOHTTP_CLIENT_SESSION_SSL", and links that spell out a port or a username, instead of rejecting them. [#34482](https://github.com/open-webui/open-webui/pull/34482), [#34479](https://github.com/open-webui/open-webui/issues/34479), [#34462](https://github.com/open-webui/open-webui/issues/34462)
|
||||
- **Qdrant version warning.** Connecting to a Qdrant 1.20 server, as the vector database or for external Qdrant knowledge bases, no longer logs a client version warning, which servers on Qdrant 1.18 or older now get instead. [#34485](https://github.com/open-webui/open-webui/pull/34485), [#34483](https://github.com/open-webui/open-webui/issues/34483)
|
||||
- **OAuth sessions on other devices.** Signing in past "OAUTH_MAX_SESSIONS_PER_USER" now removes the stored provider sessions used least recently instead of the ones created first, so a device signed in early and still in use keeps its session, and signing out now removes that browser's stored session unless it is the person's only one for that provider. [#34586](https://github.com/open-webui/open-webui/pull/34586), [#34581](https://github.com/open-webui/open-webui/issues/34581)
|
||||
- **Sign-in kept during provider outages.** A sign-in provider that is unreachable, rate limited or failing with a server error while a user's token is refreshed no longer deletes that user's stored sign-in, so tools and connections that use the sign-in token work again on the next call instead of only after signing out and back in. [#34585](https://github.com/open-webui/open-webui/pull/34585), [#34578](https://github.com/open-webui/open-webui/issues/34578)
|
||||
- **MCP OAuth with large headers.** Adding an MCP server with OAuth 2.1 and refreshing its tokens now work when its authorization server sends a very large response header, as monday.com does, instead of failing with "Connection error". [#34084](https://github.com/open-webui/open-webui/pull/34084), [#34078](https://github.com/open-webui/open-webui/issues/34078)
|
||||
- **MCP OAuth with HS256 tokens.** Connecting an MCP server over OAuth 2.1 now works when the server signs its ID token with the client secret, as Readwise does, instead of failing with "Algorithm 'HS256' requires 'oct' key". [#34593](https://github.com/open-webui/open-webui/pull/34593), [#34504](https://github.com/open-webui/open-webui/issues/34504)
|
||||
- **MCP sign-ins with rotating tokens.** Two tool calls that refresh the same MCP sign-in at the same time no longer delete the connection with providers that hand out a new refresh token on every refresh, because the refreshes now wait for each other, across workers when Redis is used. [#34582](https://github.com/open-webui/open-webui/issues/34582)
|
||||
- **MCP tools reconnect.** When an MCP server drops its connection during a reply, for example because it restarted, the next call to one of its tools now connects again instead of every later call in that reply failing with an "MCP client is not connected" error. [#34170](https://github.com/open-webui/open-webui/issues/34170)
|
||||
- **Large MCP tool lists.** MCP servers whose tool list or a single tool result is larger than 1 MB connect and answer again, also through Tether, instead of failing with "Failed to connect to MCP server". [#34421](https://github.com/open-webui/open-webui/pull/34421), [#34201](https://github.com/open-webui/open-webui/issues/34201)
|
||||
- **Local speech-to-text on pip installs.** Local speech-to-text in Open WebUI installed with pip now transcribes again instead of failing every recording with an error about "metadata_errors". [#34280](https://github.com/open-webui/open-webui/pull/34280), [#34278](https://github.com/open-webui/open-webui/issues/34278)
|
||||
- **Terminal proxy headers.** A "TERMINAL_PROXY_HEADERS" entry written in a different letter case now replaces the terminal's own header instead of being sent next to it, and repeated headers from the terminal such as several cookies are all passed on. [#34092](https://github.com/open-webui/open-webui/issues/34092)
|
||||
- **Busy servers with session sharing.** With "DATABASE_ENABLE_SESSION_SHARING" on, a burst of page loads no longer stalls for 30 seconds and fails with a "QueuePool limit" error, because chat lists and the inbox no longer open a second database connection for their extra reads. [#34185](https://github.com/open-webui/open-webui/pull/34185), [#34184](https://github.com/open-webui/open-webui/issues/34184)
|
||||
- **Quieter search logs.** Knowledge searches and several web search engines no longer write found file names or full search results to the server log at the default level, only with "GLOBAL_LOG_LEVEL" set to "DEBUG". [#34146](https://github.com/open-webui/open-webui/pull/34146), [#34129](https://github.com/open-webui/open-webui/issues/34129)
|
||||
- **Long settings text boxes.** Long prompt templates and JSON fields in Admin Settings now stop growing at a set height and scroll, instead of stretching the page to fit all of their text. [#34569](https://github.com/open-webui/open-webui/issues/34569)
|
||||
- **Deleted accounts leave access lists.** Deleting an account now also removes it from the access lists of everything it was given access to, where it used to stay as a row showing only the deleted account's ID. [#34223](https://github.com/open-webui/open-webui/pull/34223), [#30023](https://github.com/open-webui/open-webui/discussions/30023)
|
||||
- **User list API groups.** The admin endpoint that lists all users now includes each user's groups instead of an empty list. [#34489](https://github.com/open-webui/open-webui/pull/34489), [#26156](https://github.com/open-webui/open-webui/discussions/26156)
|
||||
- **Groups need a name.** A new group can no longer be saved without a name from a tab other than General, because Save stays off until a name is typed and the server now rejects empty or blank group names. [#34518](https://github.com/open-webui/open-webui/issues/34518)
|
||||
|
||||
### Changed
|
||||
|
||||
- **Direct connections removed.** Settings > Connections and Settings > Tools are gone together with the Direct Connections and Direct Integrations switches in Admin Settings, the Direct Tool Servers permission and the "ENABLE_DIRECT_CONNECTIONS", "ENABLE_DIRECT_INTEGRATIONS" and "USER_PERMISSIONS_FEATURES_DIRECT_TOOL_SERVERS" settings, so models, tool servers and terminals that users added in their own settings no longer appear and have to be added in the Tether app instead.
|
||||
- **Terminals connections need templates.** Connections to a Terminals server no longer carry a policy in Admin Settings > Integrations, so the Policy ID, image, resource, storage, idle timeout and Terminal Contexts settings and the Refresh Terminals button are gone, the server has to be a Terminals version that supports terminal templates, and models, folders, automations and chats that used the old connection need one of the new terminals selected.
|
||||
- **Image settings become connections.** On the first start of this version, the image generation and editing settings are turned into image connections that everyone who could use them before keeps using, and from then on the image environment variables such as "IMAGE_GENERATION_ENGINE", "IMAGES_OPENAI_API_BASE_URL" and "COMFYUI_BASE_URL" are no longer read unless "ENABLE_PERSISTENT_CONFIG" is turned off.
|
||||
- **Image API model checks.** Image generation and edit requests sent through the API now run on the model they name, which has to be one of the caller's image models or the request fails with an error, instead of image generation ignoring the named model.
|
||||
- **New coding agent model IDs.** Coding agent models now get IDs made only of letters, digits, dashes and underscores, which makes pinned Cursor and Pi models work, and pins, default models and automations that point at an agent model whose name had other characters, dots included as in most Codex models, need the model picked again. [#34171](https://github.com/open-webui/open-webui/pull/34171), [#34168](https://github.com/open-webui/open-webui/issues/34168)
|
||||
- **Coding agent sessions restart once.** The first new message in a coding agent chat from an earlier version starts a new agent session that gets the chat so far as text, instead of resuming the agent's own session.
|
||||
- **Attachments menu renamed.** The Add Context entry in the + menu of the message input, which holds the options to attach web pages, files, notes, knowledge and chats, is now called Attachments. [#34431](https://github.com/open-webui/open-webui/pull/34431)
|
||||
- **Terminal uploads go to files.** New terminal connections now start with Chat Uploads set to Filesystem and computers connected through Tether always use it, so files attached in a chat are saved into the selected terminal's working folder instead of being processed as regular chat uploads, while existing connections keep their setting.
|
||||
- **Live update events renamed.** Real-time updates for notes, comments, the inbox, access changes, encrypted channels and channel app errors now arrive as typed messages on the main "events" socket channel, so custom clients that listened for the old event names such as "events:note" or "note:presence" need to listen there instead.
|
||||
- **Chat write access field.** Chat responses from the API now carry the saved sharing mode in "share_mode" and report whether you can reply in a new "write_access" field, so custom clients that read "share_mode" for that need to switch.
|
||||
|
||||
## [0.13.3] - 2026-11-08
|
||||
|
||||
### Added
|
||||
|
||||
Reference in New Issue
Block a user