---
title: "Version 5.21"
slug: "ca-version-521"
updated: 2025-11-24T08:53:01Z
published: 2025-11-24T08:53:01Z
canonical: "docs.knovvu.com/ca-version-521"
---

> ## Documentation Index
> Fetch the complete documentation index at: https://docs.knovvu.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Version 5.21

## Version 5.21
### November 19, 2025
:::(Warning) (Known Issue)
Generative AI features of the solution—including Ask GenAI, GenAI-based categories, auto-tagging, translation, and AI-generated categories—
**currently do not support GPT-5.**
:::


## Enhancements

**Export Users as CSV**  
A new export option allows administrators to download all users in CSV format, simplifying bulk user management and reporting processes.

**Top 10 / Worst 10 Agents in Coaching Dashboard**  
The Coaching Dashboard now highlights the Top 10 and Worst 10 performing agents, enabling supervisors to identify strengths and weaknesses more quickly.

**Language Filter**  
A language filter has been added, allowing users to narrow down conversations or data by language for more targeted analysis.

**Display Audio Compression Settings**  
The system now retrieves and displays the audio compression bitrate and format directly from configuration settings, ensuring transparency and easier troubleshooting.

**Multiple Evaluations from Conversation Detail**  
Users can now initiate multiple evaluations for the same conversation from the detail page. To enable this feature, activate the relevant configuration under AQM settings.

**Improved Assignment Creation**  
Evaluator and agent lists are now filtered more intelligently during assignment creation, speeding up workflow and reducing selection errors.

**Redaction Support for Chat Conversations**  
Chat-type conversations are now included in the redaction/masking process, ensuring consistent data protection across all communication types.

**Dynamic LLM Provider Update**  
The LLM provider has been migrated from a product-based model to a service-based model. New LLM models are now added to CA automatically—without requiring a new release.

**Disable Excel Export for 'Not in Reports' Category Filter**  
To prevent confusion and unnecessary export operations, Excel export is now disabled when the “Not in Reports” category filter is selected.

**Quick Search for All Conversation Types**  
Quick Search functionality has been expanded to support *all* conversation types, improving discoverability and navigation throughout the system.

**Matching Categories for Email Conversations**  
Email conversations now display matching categories detected by the categorization engine, ensuring consistency with voice and chat channels.

**Custom Retention Period for Genesys–CA Integration**  
A new option allows custom retention periods specifically for Genesys-integrated conversations, enabling more granular data lifecycle control.

**Call Number Search Helper with Wildcard Support**  
A helper tool has been added for searching call numbers, including support for the `*` wildcard to simplify complex searches.

**Category Details in AQM Evaluation Forms**  
Evaluation forms now display detailed category information, allowing evaluators to better assess agent performance based on detected insights.

## Improvements


**Support for Special Characters in Regex Entities**  
Regex-based entities now support a wider range of special characters, improving accuracy in data extraction and masking rules.

**Updated Sentiment Colors for Improved UX**  
Sentiment colors have been refreshed for better visual clarity and a more intuitive user experience.

**Forced Full String/Word Masking for Regex-Based Entities**  
A new configuration allows full-string or full-word masking when using custom regex entities, ensuring stricter privacy for sensitive patterns.

**Language Synchronization Between ixHero and CA iFrame**  
Language selections are now synchronized between ixHero and the embedded CA iFrame, ensuring a seamless multilingual experience.

**New Monitoring Metrics**  
Additional monitoring metrics have been introduced to improve system observability and operational insights.

**User-Based Feedback Report – Column Rename**  
Columns for non-call users have been renamed to better reflect their role and context within the User-Based Feedback Report.

**Prevent Repetitive Requests in Coaching Dashboard**  
Unnecessary repeated API requests between Coaching Dashboard tabs have been eliminated to improve performance and responsiveness.

**Back to Feedback Navigation Button**  
A new button allows users to return directly to the feedback list after viewing a specific feedback entry.

**Departments Column in Non-FCR Page**  
The Non-FCR page now includes department information, improving traceability of unresolved cases.

**Separated Indexing for Generative AI Category Items**  
Indexing for categories containing Generative AI items has been separated, resolving performance issues previously observed with GenAI Yes/No logic.

**MinIO Upgrade**  
MinIO has been upgraded to a newer version, improving storage stability, performance, and compatibility with CA services.

**Minor UI Enhancements**
Several minor improvements have been implemented to enhance the user interface. These enhancements make the interface more intuitive and user-friendly, contributing to a better overall user experience.

**Minor Bug Fixes**
Various minor bugs have been addressed to improve overall system stability and performance. These fixes result in a smoother and more reliable user experience.