Releases 2026-03

This page shows items released in 2026-03.

Monday, March 30, 2026 - Version 2026.1.13.0 AZURE-EU AZURE-US SWITZERLAND

NEW New Connector Endpoint to Cancel Document Workflows Remotely

A new connector endpoint allows integrated external systems to cancel all jobs and tasks associated with a specific document workflow. This eliminates the need to switch between platforms to stop unnecessary processing, saving time and preventing wasted translation costs on obsolete tasks. The cancellation works for both Standard and Codyt projects, covering all active and completed job statuses (except archived ones).

NEW Default Setting for Context-Based Duplicates in Translation Memory Settings

The general translation memory settings now include a default option for "Allow context-based duplicates." This lets project managers define the duplicate handling behavior once at the translation memory level, instead of configuring it manually for every consolidation. The setting applies automatically as the default for future consolidations into that translation memory, and can still be overridden on a per-consolidation basis when needed. For more information, see: https://help.wordbee.com/wbt/translation-memory-default-search-settings

NEW Attribute Value Filtering for HTML File Translation Rules

The HTML file filter now supports filtering translatable content based on attribute values, not just attribute names. This gives you more precise control over which parts of your HTML files are sent for translation. For more information, see: https://help.wordbee.com/wbt/web-page-configuration-options

IMPROVED Whitespace Compression Option for DITA File Processing

A new whitespace compression option has been added to the DITA file filter. When enabled, consecutive whitespace characters in extracted content are collapsed into a single space, helping to keep translations clean and consistent. For more information, see: https://help.wordbee.com/wbt/dita-files

IMPROVED QA Report Now Includes Source and Target Content

The QA Issues report template now includes the source and target text alongside each reported issue. Additionally, the Report module in the editor now automatically preselects the target language based on your current cursor position, making it faster to generate reports for the language you are working on.

IMPROVED Clearer Labels for Resource Usage in MT Hive Order Form

The MT Hive order form settings now include clearer descriptions explaining which translation memories are used for pretranslation versus post-editing. This helps users understand that pretranslation resources are only used for the initial translation step, while post-editing project resources come from the configured template project. For more information, see: https://help.wordbee.com/wbt/how-to-configure-and-assign-mt-hive-forms

IMPROVED DITA Files Now Respect Non-Translatable Content Markers

DITA files that mark certain sections as non-translatable are now handled correctly. Content flagged as not requiring translation is automatically excluded from the extraction process, along with all nested content within those sections. For more information, see: https://help.wordbee.com/wbt/dita-files

IMPROVED Email Notifications Now Always Use Configured Sender Address

All email notifications, including multi-factor authentication emails, now consistently use the sender email address configured in your platform settings. Previously, some emails could be sent from a different address than the one you configured, which was particularly problematic for platforms using custom branding. This ensures recipients always see a consistent, trusted sender across all notification types.

IMPROVED Reference Material Auto-Copy for Order-Based Projects

The automatic reference material copy feature, which copies files from a client's Templates folder to new projects, now also works for projects created from orders. Previously, this only worked for manually created projects, requiring users to upload reference files by hand for order-based workflows.

IMPROVED Order Notifications Now Correctly Follow Role-Based Settings

Order notifications are now based on each participant's actual role in the order, submitter/co-owner or manager, rather than on which company they belong to. Previously, participants could be incorrectly classified, causing them to miss notifications entirely or receive them under the wrong category. For example, when a co-owner from the master company sent an order message, the order manager might not be notified, or when an outsourcing supplier managing the order sent a message, the submitter might not receive it. Notifications now reliably match the two columns shown in notification settings: "Orders submitted/co-owned by user" and "Orders managed by user".

IMPROVED Faster Supplier Finder Loading for Large Supplier Networks

The Supplier Finder assignment window now loads significantly faster for platforms with extensive supplier networks. This optimization ensures a smoother and more responsive experience when browsing and assigning suppliers to jobs.

IMPROVED Parallel Processing for Translate Files Requests

A new optional parameter allows multiple Translate Files requests to run at the same time for the same user, instead of being processed one after another. This can significantly reduce total processing time for high-volume translation workflows. By default, the previous sequential behavior is unchanged. For more information, see: https://help.wordbee.com/api2/apps-tools-translatefiles-translate-post

IMPROVED Automatic Cleanup of Empty Folders After File Deletion

When files are deleted from the Document Library, any folders that become empty as a result are now automatically removed. This cleanup works recursively, so if removing a folder leaves its parent folder empty, that parent folder is also removed. This keeps the Document Library clean and organized without requiring manual folder cleanup.

FIX Create Segment Popup Now Fully Editable for Termbase Resources

Fixed a bug where the Create Segment popup in the editor opened in read-only mode when working with Termbase resources on platforms without the Terminology Management tool enabled. The popup is now fully functional regardless of the platform configuration.

FIX Fixed File Processing Failure for Files with Large Character Constraints

Fixed an issue where XLIFF files with very large character length constraints (exceeding system limits) would fail to process. The system now handles these values gracefully, ensuring files are imported and indexed correctly.

FIX Fixed Word Count Inconsistency Between Target and V2 Target Languages in Codyt Projects

In Codyt projects that use versioned target languages, such as when a back-translation workflow adds a v2 version of an existing target language, the word count analysis could produce different results for the original target and its v2 version, even when both shared the same source content and translation memory settings. This inconsistency affected costing and invoicing calculations. The issue has been fixed so that both language versions now produce identical word counts when all configuration parameters are the same.

FIX Fix for Translate Files ZIP Download Failure

Fixed an issue where downloading translated ZIP files from the Translate Files tool would fail with a "File not found" error. Single-file uploads were not affected.

FIX Fixed Order Form Settings Page Not Loading for Non-Admin Users

On some platforms, non-admin users with the appropriate access right to manage order forms could not open the settings page for certain older order forms. The page would fail to load when a custom field was not accessible to the user. This has been fixed so that the settings page now loads correctly for all order forms, regardless of custom field availability.

FIX Document header row disappears when scrolling up and down the page while having infinite scrolling on

Document header row disappears when scrolling up and down the page while having infinite scrolling on Document header row disappears when scrolling up and down the page while having infinite scrolling on

FIX Custom Field Placeholders Now Populate Consistently in New Order Form

Fixed an issue where project-level custom field values were not always populated when creating a new order through the order form. Custom field placeholders now behave consistently regardless of the order creation method.

FIX Fixed Deep Link Navigation After Single Sign-On Login

When logging in via Single Sign-On through a deep link to a specific document, the system now correctly opens the intended document. Previously, only the first part of the link was preserved after login, causing the editor to open all documents instead of the specific one linked.

Tuesday, March 3, 2026 - Version 2026.1.10.0 AZURE-EU AZURE-US SWITZERLAND

NEW Multiple Job Assignment to the Same Supplier for Codyt

You can now select multiple Codyt jobs with different language pairs and task types, and assign them all to the same supplier in a single action. Previously, each language pair had to be assigned separately, which was time-consuming for projects with many languages. The redesigned "Assign jobs" screen offers a new "Select supplier" mode where you choose a supplier, optionally set a specific worker, and apply the assignment across all selected jobs at once.

IMPROVED Client Portal Now Uses Centralized Branding from TMS Settings

The Client Portal (MT Hive) now automatically uses the logos, banners, and visual assets configured in your TMS "Look & Feel" settings. This means you only need to manage your branding in one place, and it will be consistently applied across the entire platform. Existing custom branding on the Client Portal is preserved and not affected by this change.

IMPROVED Improved Description for R200 Access Right to Reflect its Full Scope

Updated the description of access right R200 (Can search and access projects) to clearly explain its full scope. The new description states that R200 controls access to the Projects section and all project-related data, and that it is also required to view client quotes and invoices across the platform, including from the order details page. This helps administrators assign permissions more confidently and avoid unexpected access gaps.

IMPROVED Added Invoice Tracking to Admin Traces API

Invoices are now tracked in the admin audit trail via the Traces API (admin/logs/traces). When an invoice is created, modified, or deleted, the action is recorded along with who performed it and when. This allows administrators to investigate invoice changes, such as accidental deletions, through the existing traces method.

IMPROVED Forbidden Terms QA Rule Now Supports Chinese, Japanese, and Korean Languages

The Forbidden Terms QA rule now correctly detects forbidden terms in Chinese, Japanese, and Korean content. Previously, the rule could miss forbidden terms in these languages because they do not use spaces between words. The rule now automatically applies the appropriate matching method for these languages, ensuring reliable quality checks across all supported languages.

IMPROVED Improved Performance for Translation Memory Consolidation with Large Datasets

The consolidation process that copies and merges segments from projects into translation memories has been optimized for better performance with large datasets. Memory usage during consolidation is now significantly reduced, preventing potential slowdowns or failures when processing translation memories with a large number of segments. This ensures more reliable and stable consolidation operations for enterprise-scale translation memories.

IMPROVED Job Reference Column is Now Optional in the Jobs List

The "Job Reference" column in the jobs list is no longer mandatory. You can now hide or show it based on your preference, giving you more control over your workspace layout. This is especially useful if you primarily use the Job ID to identify jobs and want a cleaner, less cluttered view.

IMPROVED Translate Files API Now Supports ZIP Multi-file Processing and MT Consolidation

The Translate Files API endpoint (/api/apps/tools/translatefiles/translate) now supports submitting up to 100 files at once via ZIP archives, reducing the number of API calls needed for bulk operations. Additionally, a new consolidateResourceId parameter enables MT consolidation, automatically saving machine-translated segments into a Translation Memory so that future submissions can reuse prior results and reduce MT costs. These features are designed for large-scale, high-volume translation workflows.

IMPROVED Unique Error Codes for Concurrent License Failures in Login Reports

The Failed Login Attempts report now distinguishes between login failures caused by seat-limit restrictions and incorrect password attempts. When a Power User or Linguist is blocked because all available seats are occupied, the report displays a specific error code and message indicating which user type has reached its limit. This makes it easier for administrators to identify capacity issues and take action, such as ordering additional licenses.

FIX Fixed API Throttle Counting for Multi-file Upload

Fixed an issue where the multi-file upload API endpoint incorrectly counted each uploaded file as a separate API call against the throttling limit. A single multi-file upload request now correctly counts as one API call regardless of the number of files included, preventing users from unexpectedly hitting their API rate limits during bulk uploads.

FIX Fixed TM Hits Display in Excel Export from Editor

Fixed an issue where Translation Memory (TM) hits exported to Excel from the Editor appeared as unreadable raw data instead of properly formatted text. TM hits in Excel exports now display with the same readable formatting as in Word exports, with source and target text clearly separated and color-coded.

FIX Business Analytics Data Loss

The Business Analytics engine could crash when calculating job statistics if a database record was being modified at the same time by another user, such as someone editing a document while analytics were running. This caused the entire day's data for the affected account to stop processing. Additionally, failed days were incorrectly marked as complete, meaning the data was permanently skipped and never retried. The fix ensures failed calculations are retried automatically and individual failures no longer block the rest of the data from being processed.

FIX Excel Parser Markup Transfer to Context/Key Column

Resolved an issue where formatting markup from source text in Excel files was incorrectly appearing in the context/key column when using context/key column extraction with the "Content is html" option enabled. The markup was also being removed from the source text where it should have remained. The context/key column now correctly displays only the value from the designated column, and formatting in the source text is preserved as expected.

FIX Typo Corrected for Arabic Language Label in French

We have corrected a spelling error in the language list under Customization > Languages, where "Arabe" was incorrectly displayed as "Aarabe" for French.

FIX Incorrect Requester Shown on Job Status Popup

In projects linked to multiple client orders, the job status popup was displaying the requester and client company from the most recent order linked to the project, instead of the order that actually created the job. The popup now correctly shows the requester and company from the specific order that generated the job.