Skip to content

Releases: MindWorkAI/AI-Studio

Release list

Release v26.7.2

Release v26.7.2 Latest
Latest

Choose a tag to compare

View all tags
@github-actions github-actions released this 06 Jul 19:07
v26.7.2
331b8f3
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Verified
Learn about vigilant mode.

v26.7.2, build 244 (2026-07-06 18:35 UTC)

  • Improved the coding assistant with a simpler context workflow. Now you can attach multiple source files directly instead of manually defining code contexts.
  • Improved the grammar & spelling checking and rewrite & improve text assistants. Both assistants can now load one document into the text field using the file dialog or drag and drop.
  • Improved the legal check, text summarizer, and translation assistant. These assistants can now load documents with drag and drop.
  • Fixed the assistant builder so it can continue generating an assistant in the background if you leave the page and come back later.
  • Fixed text fields in assistants added through assistant plugins so they now respect the app's spell checking settings.
  • Fixed the dialog for adding providers. Selecting an option could previously prevent the setup from continuing. Thanks, Dominic Neuburg (donework), for reporting this issue.

Virus scans

Assets 21

Release v26.7.1

Release v26.7.1

Choose a tag to compare

View all tags
@github-actions github-actions released this 05 Jul 17:13
v26.7.1
5ea2d35
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Verified
Learn about vigilant mode.

v26.7.1, build 243 (2026-07-05 16:39 UTC)

  • Added the assistant builder as a beta preview feature for generating and installing assistant plugins. This no-code environment is designed so anyone can create assistants without needing a computer science background. To try it, enable preview features in the app settings and then explicitly enable the Assistant Builder feature. Thanks, Nils Kruthoff (nilskruthoff), for this extraordinary contribution.
  • Added expert capability overrides for providers, so advanced users and configuration plugins can manually adjust selected model capabilities, including reasoning (thinking) behavior, when automatic detection needs adjustment.
  • Added configuration plugin options for default chat data source behavior and the related data source selection and validation agents.
  • Added support for organization-approved assistant plugins, so trusted assistant plugins can be enabled without requiring each user to run a separate security audit.
  • Added support for assistant plugin tiles that can open a chat directly in a chosen workspace.
  • Improved the provider selection by showing small capability icons for supported audio, image, speech, and reasoning features of the selected model.
  • Improved PDF file handling in Flatpak builds by supporting the standard Flatpak library location for PDFium and adding clearer diagnostics for troubleshooting PDF loading.
  • Improved source links in chat answers when file or document names contained spaces, umlauts, or other special characters. Source entries now open much more reliably for documents with names such as PDFs from shared portals or internal knowledge bases.
  • Improved all assistants, so running tasks can continue when you leave the assistant and return later.
  • Improved the assistant overview so it shows which assistants are still running or have a result ready.
  • Improved the activity indicators for running chats and assistants so they use a consistent blue highlight.
  • Improved the chat experience by automatically focusing the message composer again when it becomes available. Thanks, Dominic Neuburg (donework), for the contribution.
  • Improved the chat message formatting so line breaks in responses are preserved more closely, making structured text easier to read. Thanks, Dominic Neuburg (donework), for the contribution.
  • Fixed configuration plugins so they can configure assistant plugin audit settings, including required audit levels, audit provider selection, and automatic background audits.
  • Fixed default data source selection so sources that require local or trusted providers are selected again after switching from a cloud provider to a suitable provider.
  • Upgraded Rust to v1.96.1

Virus scans

Loading

Release v26.6.2

Release v26.6.2

Choose a tag to compare

View all tags
@github-actions github-actions released this 21 Jun 17:22
v26.6.2
5af616f
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Verified
Learn about vigilant mode.

v26.6.2, build 242 (2026-06-21 14:07 UTC)

  • Added a read-only view for organization-managed profiles and chat templates, so users can inspect the content while the organization remains in control of changes.
  • Added support for organization-managed chat defaults. Configuration plugins can now preselect the chat provider, profile, and chat template, either as locked values or editable defaults.
  • Added support for organization-managed introduction texts on the home page. Configuration plugins can now add custom Markdown introductions and hide the built-in introduction. Thanks, Harald, for the feedback about the built-in introduction text.
  • Added support for organization-managed provider confidence settings. Configuration plugins can now set confidence presets, custom confidence schemes, and an app-wide minimum confidence level.
  • Added support for organization-trusted providers in data source security checks. Configuration plugins can now mark specific provider instances as trusted for data source usage and local embedding warnings.
  • Changed provider confidence settings to appear in their own settings panel, because they apply to LLM, embedding, and transcription providers.
  • Fixed chat provider, profile, and template selections not updating live after configuration plugins were changed.
  • Fixed organization-managed chat templates not showing the correct icon in the chat template selection menu.
  • Fixed personal settings sometimes being lost after a settings-format upgrade when an older app version was started again. AI Studio now keeps versioned settings backups, restores the latest compatible backup when needed, and warns users when settings cannot be saved safely.
  • Fixed self-hosted provider API keys sometimes being stored under a localized name. AI Studio now uses a stable key name, keeps correct entries working, and automatically migrates known localized entries for LLM, transcription, and embedding providers. Organizations using configuration plugins do not need to change their plugins; affected users who still see an invalid API key warning should open the provider, transcription, or embedding settings and update the API key once. Thanks, Tim & Eric, for the detailed bug report and testing help.

Virus scans

Loading

Release v26.6.1

Release v26.6.1

Choose a tag to compare

View all tags
@github-actions github-actions released this 11 Jun 14:37
v26.6.1
4c328c8
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Verified
Learn about vigilant mode.

v26.6.1, build 241 (2026-06-11 13:49 UTC)

  • Added support for up to 100 thousand enterprise configuration slots, using fixed-width slot names such as config_00000 while keeping the existing first ten slot names compatible.
  • Added an enterprise configuration option to hide the quick start guide on the welcome page.
  • Added support for managed custom root certificate bundles and host allowlists for external HTTPS requests, helping Flatpak deployments connect to organization-internal services with private root CAs while keeping built-in cloud provider endpoints on system trust.
  • Added support for reading enterprise policy files from a Flatpak provisioning extension.
  • Added startup path and Linux package type details to the information page to make support easier.
  • Added the option to search for chats in all workspaces.
  • Improved self-hosted llama.cpp providers by loading available models from the server and supporting servers that offer multiple models. Thanks to the GONICUS team for reporting this issue.
  • Improved workspaces by highlighting the currently open chat in the workspace view.
  • Improved workspaces by adding a shortcut to start a new chat directly from each workspace row.
  • Improved workspaces by allowing new workspaces to be created while moving a chat.
  • Improved voice recording shortcut labels so they match the user's keyboard layout after being configured.
  • Improved the enterprise configuration details on the information page by showing where each configuration comes from and which configuration slot was used.
  • Fixed an issue where newly added profiles and chat templates were not usable until the app was restarted.
  • Fixed an issue where renamed chat templates and profiles continued to show their old names in the chat toolbar until the app was restarted.
  • Fixed workspace creation and renaming to prevent new workspaces from using an existing name.
  • Fixed an issue on Microsoft Windows where reading attached documents could briefly open a terminal window while processing files.
  • Fixed an issue where AI Studio could be started multiple times on Microsoft Windows by launching it from different virtual desktops.
  • Fixed an issue where Flatpak installations could not find Pandoc from the bundled plugin extension.
  • Upgraded dependencies.

Virus scans

Loading

Release v26.5.5

Release v26.5.5

Choose a tag to compare

View all tags
@github-actions github-actions released this 25 May 19:23
v26.5.5
9f18a50
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Verified
Learn about vigilant mode.

v26.5.5, build 240 (2026-05-25 18:52 UTC)

  • Released the voice recording and transcription feature for all users. You no longer need to enable a preview feature to configure transcription providers, select a transcription provider, or use dictation.
  • Added export options for profiles and chat templates, including an option to package chat template attachments into configuration plugins.
  • Added support for organization-managed ERI servers in configuration plugins, so admins can preconfigure external data sources for users.
  • Added an export option for ERI server data sources, so admins can create configuration plugin snippets without writing the Lua code manually.
  • Added an option to configure the timeout setting for all requests. This is useful when you have a slow network connection, or you have to work with slow AI servers. It is also possible to configure this timeout for an entire organization using configuration plugins.
  • Added the ability to keep multiple chats running at the same time. For example, you can let a complex research chat continue in the background while you use some assistants to improve a text.
  • Added the username to the information page to make organization support easier when users share their screen.
  • Improved the app's security foundation with major modernization of the native runtime and its internal communication layer. This work is mostly invisible during everyday use, but it replaces older components that no longer received the security updates we require. We also continued updating security-sensitive dependencies so AI Studio stays on a healthier, better maintained base.
  • Improved the Pandoc management and detection process to make it more reliable.
  • Improved the Qdrant startup and vector database initialization, so AI Studio can start more reliably while the local vector database is still starting.
  • Fixed the Pandoc installation, which could fail and prevent AI Studio from installing its local Pandoc dependency.
  • Fixed an issue where the spellchecking setting was not applied to all text fields in the slide builder assistant.
  • Fixed an issue where legacy .doc files could be selected even though AI Studio could not process them. These files are now rejected with a clear error message. Thanks to Bernhard for reporting this issue.
  • Fixed an issue where attached documents were detached when editing a previous prompt. They now remain attached.
  • Fixed an issue where failed transcription requests could be shown as empty transcription results instead of a clear error message.
  • Fixed an issue where an AI response in chat could be interrupted when you interacted with workspaces, such as opening, closing, or resizing the workspace panel.
  • Fixed an issue with switching between chat threads while multiple chats are running.
  • Fixed error messages for provider requests so missing OpenAI API credits and too many requests are shown clearly in chats, assistants, transcription, and model loading.
  • Fixed missing translations for file type names in file selection dialogs.
  • Fixed the chat user prompt being cleared when toggling the workspace view.
  • Upgraded the native secret storage integration to keyring-core, keeping API keys in the secure credential store provided by the operating system.
  • Upgraded Rust to v1.95.0.
  • Upgraded .NET to v9.0.16.
  • Upgraded Tauri to v2.11.1. Thanks to Paul PaulKoudelka for working on this migration.
  • Upgraded PDFium to v148.0.7763.0.
  • Upgraded Qdrant to v1.18.1.
  • Upgraded other dependencies as well.

Virus scans

Loading

Release v26.4.1

Release v26.4.1

Choose a tag to compare

View all tags
@github-actions github-actions released this 17 Apr 17:58
v26.4.1
519abe4
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Verified
Learn about vigilant mode.

v26.4.1, build 235 (2026-04-17 17:25 UTC)

  • Added support for the latest AI models, e.g., Qwen 3.5 & 3.6-family, Mistral Large 3 & Small 4, OpenAI GPT 5.4, Claude Opus 4.7 etc.
  • Added assistant plugins, making it possible to extend AI Studio with custom assistants. Many thanks to Nils Kruthof nilskruthoff for this contribution.
  • Added a slide planner assistant, which helps you turn longer texts or documents into clear, structured presentation slides. Many thanks to Sabrina Sabrina-devops for her wonderful work on this assistant.
  • Added a reminder in chats and assistants that LLMs can make mistakes, helping you double-check important information more easily.
  • Added the ability to format your user prompt in the chat using icons instead of typing Markdown directly.
  • Added the ability to load a system prompt from a file when creating or editing chat templates.
  • Added a prompt optimization assistant that helps you create more effective prompts.
  • Added a start-page setting, so AI Studio can now open directly on your preferred page when the app starts. Configuration plugins can also provide and optionally lock this default for organizations.
  • Added pre-call validation to check if the selected model exists for the provider before making the request.
  • Added math rendering in chats for LaTeX display formulas, including block formats such as $$ ... $$ and \[ ... \].
  • Added support for mandatory information notices in configuration plugins. Organizations can now require users to read and confirm important information before continuing in AI Studio.
  • Released the document analysis assistant after an intense testing phase.
  • Improved enterprise deployment for organizations: administrators can now provide up to 10 centrally managed enterprise configuration slots, use policy files on Linux and macOS, and continue using older configuration formats as a fallback during migration.
  • Improved transparency on the information page by showing configured mandatory notices together with their source and confirmation status.
  • Improved the profile selection for assistants and the chat. You can now explicitly choose between the app default profile, no profile, or a specific profile.
  • Improved the performance by caching the OS language detection and requesting the user language only once per app start.
  • Improved the chat performance by reducing unnecessary UI updates, making chats smoother and more responsive, especially in longer conversations.
  • Improved the information page so enterprise configuration details now refresh live when your organization's configuration changes.
  • Improved the workspace loading experience: when opening the chat for the first time, your workspaces now appear faster and load step by step in the background, with placeholder rows so the app feels responsive right away.
  • Improved the reliability of the global voice recording shortcut so it stays available more consistently.
  • Improved the user-language logging by limiting language detection logs to a single entry per app start.
  • Improved the logbook readability by removing non-readable special characters from log entries.
  • Improved the logbook reliability by significantly reducing duplicate log entries.
  • Improved file attachments in chats: configuration and project files such as Dockerfile, Caddyfile, Makefile, or Jenkinsfile are now included more reliably when you send them to the AI.
  • Improved the validation of additional API parameters in the advanced provider settings to help catch formatting mistakes earlier.
  • Improved the model checks and model list loading by showing clearer error messages when AI Studio cannot access a provider because the API key is missing, invalid, expired, or lacks the required permissions.
  • Improved the app startup resilience by allowing AI Studio to continue without Qdrant if it fails to initialize.
  • Improved the translation assistant by updating the system and user prompts.
  • Improved how results from assistants are transferred into chats, so follow-up conversations now show clearer context while keeping the chat easier to understand.
  • Improved OpenAI-compatible providers by refactoring their streaming request handling to be more consistent and reliable.
  • Improved the app update check so malformed update metadata is handled more reliably and users now receive a clear error message instead of a misleading "No update found" response.
  • Fixed an issue where assistants hidden via configuration plugins still appear in "Send to ..." menus. Thanks, Gunnar, for reporting this issue.
  • Fixed an issue with chat templates that could stop working because the stored validation result for attached files was reused. AI Studio now checks attached files again when you use a chat template.
  • Fixed an issue with voice recording where AI Studio could log errors and keep the feature available even though required parts failed to initialize. Voice recording is now disabled automatically for the current session in that case.
  • Fixed an issue where the app could turn white or appear invisible in certain chats after HTML-like content was shown. Thanks, Inga, for reporting this issue and providing some context on how to reproduce it.
  • Fixed an issue where file and folder selection dialogs could open more than once on Windows. Thanks to Bernhard for reporting this bug.
  • Fixed an issue where exporting to Word could fail when the message contained certain formatting.
  • Fixed security issues in the native app runtime by strengthening how AI Studio creates and protects the secret values used for its internal secure connection.
  • Updated several security-sensitive Rust dependencies in the native runtime to address known vulnerabilities.
  • Updated .NET to v9.0.15
  • Updated dependencies

Virus scans

Loading

Release v26.2.2

Release v26.2.2

Choose a tag to compare

View all tags
@github-actions github-actions released this 22 Feb 14:40
v26.2.2
09df19e
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Verified
Learn about vigilant mode.

v26.2.2, build 234 (2026-02-22 14:14 UTC)

  • Added a vector database (Qdrant) as a building block for our local RAG (retrieval-augmented generation) solution. Thank you very much, Paul (PaulKoudelka), for this major contribution. Note that our local RAG implementation remained in preview and has not yet been released; other building blocks are not yet ready.
  • Added an option in the embedding providers table to test the embedding process.
  • Added an app setting to enable administration options for IT staff to configure and maintain organization-wide settings.
  • Added an option to export all provider types (LLMs, embeddings, transcriptions) so you can use them in a configuration plugin. You'll be asked if you want to export the related API key too. API keys will be encrypted in the export. This feature only shows up when administration options are enabled.
  • Added an option in the app settings to create an encryption secret, which is required to encrypt values (for example, API keys) in configuration plugins. This feature only shows up when administration options are enabled.
  • Added the option to set a predefined provider for the entire app via configuration plugins.
  • Added support for using multiple enterprise configurations simultaneously. Enabled organizations to apply configurations based on employee affiliations, such as departments and working groups. See the enterprise configuration documentation for details.
  • Added the DEPLOYED_USING_CONFIG_SERVER field for configuration plugins so enterprise-managed plugins can be identified explicitly. Administrators should update their configuration plugins accordingly. See the enterprise configuration documentation for details.
  • Improved the enterprise configuration synchronization to be fail-safe on unstable or unavailable internet connections (for example, during business travel). If metadata checks or downloads fail, AI Studio keeps the current configuration plugins unchanged.
  • Improved the document analysis assistant (in beta) by hiding the export functionality by default. Enable the administration options in the app settings to show and use the export functionality. This streamlines the usage for regular users.
  • Improved the workspaces experience by using a different color for the delete button to avoid confusion.
  • Improved single-input dialogs (e.g., renaming chats) so pressing Enter confirmed immediately and the input field focused automatically when the dialog opened.
  • Improved the plugins page by adding an action to open the plugin source link. The action opens website URLs in an external browser, supports mailto: links for direct email composition.
  • Improved the configuration plugins by making EnabledPreviewFeatures additive rather than exclusive. Users can now enable additional preview features without being restricted to those selected by the configuration plugin.
  • Improved the system language detection for locale values such as C and variants like de_DE.UTF-8, enabling AI Studio to apply the matching UI language more reliably.
  • Fixed an issue where leftover enterprise configuration plugins could remain active after organizational assignment changes during longer absences (for example, vacation), which could lead to configuration conflicts.
  • Fixed an issue where manually saving chats in workspace manual-storage mode could appear unreliable during response streaming. The save button is now disabled while streaming to prevent partial saves.
  • Fixed an issue where in some places "No profile" was displayed instead of the localized text.
  • Fixed a bug in the Responses API of our OpenAI provider implementation where streamed whitespace chunks were discarded. We thank Oliver Kunc OliverKunc for his first contribution in resolving this issue. We appreciate your help, Oliver.
  • Fixed a bug in the Microsoft Word export via Pandoc where target paths containing spaces or Unicode characters were split into invalid command arguments, causing export failures. Thanks to Bernhard for reporting this issue.
  • Fixed the Google Gemini model API. Switched to the default OpenAI-compatible API to retrieve the model list after Google changed the previous API, which stopped working.
  • Upgraded to .NET 9.0.13 & Rust 1.93.1.
  • Upgraded dependencies.

Virus scans

Loading

Release v26.2.1

Release v26.2.1

Choose a tag to compare

View all tags
@github-actions github-actions released this 01 Feb 19:40
v26.2.1
e65874d
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Verified
Learn about vigilant mode.

v26.2.1, build 233 (2026-02-01 19:16 UTC)

  • Added the ability to individually configure the minimum confidence level, standard profile, and default provider for each policy in the Document Analysis Assistant (in preview).
  • Added support for defining document analysis policies (in preview) by configuration plugins, enabling centralized management of policies across entire departments or organizations.
  • Added an option to hide the policy definition in the Document Analysis Assistant (in preview) when exporting and distributing that policy by a configuration plugin in organizations, making it easier for users to use.
  • Added the policy export functionality to the Document Analysis Assistant (in preview). You can now export policies as Lua code for a configuration plugin to distribute the policy across your organization.
  • Improved the error checking & logging behavior when the installed PDFium version did not meet the minimum required version.
  • Improved the logging behavior when parsing profile tables from configuration plugins.
  • Fixed a bug where the global minimum confidence level was not being applied to the assistants.
  • Upgraded the Document Analysis Assistant and the voice recording with transcription feature (both in preview) from the prototype to the beta state. Both features are now completely implemented and are undergoing a deeper testing phase in preparation for release.

Virus scans

Loading

Release v26.1.2

Release v26.1.2

Choose a tag to compare

View all tags
@github-actions github-actions released this 25 Jan 14:29
v26.1.2
09f93c6
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Verified
Learn about vigilant mode.

v26.1.2, build 232 (2026-01-25 14:05 UTC)

  • Added the option to hide specific assistants by configuration plugins. This is useful for enterprise environments in organizations.
  • Added the current date and time to the system prompt for better context in conversations. Thanks Peer peerschuett for the contribution.
  • Added the ability to control the voice recording with transcription (in preview) by using a system-wide shortcut. The shortcut can be configured in the application settings or by using a configuration plugin. Thus, a uniform shortcut can be defined for an entire organization.
  • Added error handling for the context window overflow, which can occur with huge file attachments in chats or the document analysis assistant.
  • Added Rust failure detection to the .NET server. This is helpful in order to handle critical failures and shut down the application gracefully.
  • Improved the error handling for model loading in provider dialogs (LLMs, embeddings, transcriptions).
  • Improved the microphone handling (voice recording & transcription preview) so that all sound effects and the voice recording are processed without interruption.
  • Improved the handling of self-hosted providers in the configuration dialogs (LLMs, embeddings, and transcriptions) when the host cannot provide a list of models.
  • Improved the document analysis assistant (in preview) by allowing users to send results to a new chat to ask follow-up questions. Thanks to Sabrina Sabrina-devops for this contribution.
  • Improved the developer experience by detecting incorrect CPU architecture metadata when checking and installing the Pandoc dependency.
  • Improved the error messages for failed communication with AI servers.
  • Improved the error handling for the enterprise environment service regarding the communication with our Rust layer.
  • Fixed a bug in the document analysis assistant (in preview) where the first section containing the policy definition did not render correctly in certain cases.
  • Fixed a logging bug that prevented log events from being recorded in some cases.
  • Fixed a bug that allowed adding a provider (LLM, embedding, or transcription) without selecting a model.
  • Fixed a bug with local transcription providers (voice recording & transcription preview) by handling errors correctly when the local provider is unavailable.
  • Fixed a bug with local transcription providers (voice recording & transcription preview) by correctly handling empty model IDs.
  • Fixed a bug affecting the transcription preview: previously, when you stopped music or other media, recorded or dictated text, and then tried to resume playback, the media wouldn’t resume as expected. This behavior is now fixed.
  • Fixed a rare bug that occurred when multiple threads tried to manage the same chat thread. Thanks Sabrina Sabrina-devops for identifying and reporting this issue.
  • Fixed a bug that prevented text localization from certain source code files under specific conditions.
  • Upgraded to Rust 1.93.0
  • Upgraded to .NET 9.0.12
  • Upgraded dependencies

Virus scans

Loading

Release v26.1.1

Release v26.1.1

Choose a tag to compare

View all tags
@github-actions github-actions released this 11 Jan 16:21
v26.1.1
ab76180
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Verified
Learn about vigilant mode.

v26.1.1, build 231 (2026-01-11 15:53 UTC)

  • Added the option to attach files, including images, to chat templates. You can also define templates with file attachments through a configuration plugin. These file attachments aren’t copied—they’re re-read every time. That means the AI will pick up any updates you make to those files.
  • Added the option to use source code files in chats and document analysis. This supports software development workflows.
  • Added a preview feature that lets you record your own voice for transcription. The feature remains in development and appears only when the preview feature is enabled.
  • Added the option to configure transcription providers (for example, using Whisper models). As usual, there can be local as well as cloud models configured. This option is part of the transcription preview and remains hidden until the preview is activated or the feature gets released. Transcription providers can be configured through configuration plugins as well.
  • Added an option to the app settings to select a provider for transcribing dictated text. If no provider is selected, dictation and text transcription are disabled.
  • Added the option to configure embedding providers through a config plugin and distribute them within an organization.
  • Improved the app versioning. Starting in 2026, each version number includes the year, followed by the month. The last digit shows the release number for that month. For example, version 26.1.1 is the first release in January 2026.
  • Fixed a bug in the profile selection where the "Use no profile" entry could not be localized, causing English text to appear in languages such as German. This behavior has now been fixed.
  • Fixed a bug in the provider dialogs (LLMs, embeddings, and transcriptions) when editing a provider. In cases where an error had to be displayed, a non-localized message in English was used.
  • Fixed a very rare bug in the provider dialogs (LLMs, embeddings, and transcriptions) where a validation error appeared if the API key could not be read from the operating system, but the error did not clear after the user changed the API key.
  • Upgraded dependencies.

Virus scans

Loading