Release Notes - What's new in Kyvos 2025.7.x

Here’s a list of Kyvos versions for Kyvos 2025.7.x releases:

• Kyvos 2025.7.x - Features and enhancements

• Kyvos 2025.7 - Features and enhancements

Features and Enhancements for 2025.7.x

The Kyvos 2025.7.x release brings the following new features and enhancements.

Kyvos Dialogs Desktop Application Support

Kyvos now offers the Kyvos Dialogs Desktop Application, designed for small and mid-sized businesses that need a fast, lightweight analytics solution. It delivers quick deployment, smooth performance, and natural language insights—without the need for extensive infrastructure.
For further details about the application, visit Desktop-Based Kyvos Dialogs.

Support for Custom Attribute Names in LDAP Configuration

Kyvos Manager now supports custom attribute names for username mapping in LDAP configuration, rather than relying solely on directory-type–specific predefined attributes. The configured custom attribute is preserved across upgrades, ensuring that users, along with their associated roles and privileges, remain intact.

Features and Enhancements for 2025.7

The Kyvos 2025.7 release brings the following new features and enhancements.

• Features and Enhancements

• Querying Enhancements

• Cloud Support Enhancements

• Kyvos Manager Enhancements

• Kyvos Viz. Enhancements

• API Enhancements

• Behavior Changes

Kyvos Modern Architecture (No-Spark Ingestion) Enhancements

This release includes the following No-Spark enhancements:

• Dedicated Compute Server Support in YARN-Based On-Premises Deployments (CDP/BDH): Kyvos now supports launching dedicated compute servers within YARN containers, enhancing compatibility with YARN-based on-premises deployments. This enhancement enables users to run dedicated compute servers specifically for semantic model processing, ensuring improved resource isolation, scalability, and performance.

• Exclude Smart Aggregates for Attributes and Levels: Users can now exclude smart aggregates for specific attributes and levels. Only base aggregates will be created, and queries will use these exclusively. This option is available when raw data querying is enabled on the semantic model.

• Support of SCD Type 1 Dimensions in Semantic Models: Kyvos now supports processing, querying, and updating aggregates for semantic models with SCD Type 1 Dimensions.

• Filters on Calculated Members in Excel Sheets: Users can now apply filters such as ‘Keep Only’ and ‘Hide’ on calculated members in Excel sheets, providing greater flexibility in data analysis.

Kyvos Dialogs Enhancements

This release comes with the following Kyvos Dialogs enhancements:

• Streamlined Feedback Prompting: Feedback is now saved at the individual user level by default. This means each user’s feedback is private and not shared across users, ensuring more personalized and accurate results.

• Share Chat Sessions via Unique Links: Users can now share chat sessions through unique URLs. Each saved session generates a distinct link shown in the browser’s address bar, making it easy to share and revisit chats directly.

• Upload Excel and CSV Files in Kyvos Dialogs: Users can now upload Excel and CSV files directly in Kyvos Dialogs, allowing them to quickly start conversational analysis using structured data files. Uploaded files can also be deleted when no longer needed.

• OAuth 2.0 Support for Azure OpenAI Connections: Kyvos now supports OAuth 2.0 authentication for establishing connections with Azure OpenAI services.

• Internationalization Support: This release supports the user’s preferred language in the Kyvos Dialogs.

Secured TLS and HTTPS Communication for Kyvos Reporting

With this release, the setup for secured (TLS and HTTPS) communication between the Kyvos Reporting Server and Kyvos Reporting Portal is now automated. Users can configure these settings directly via the Kyvos Reporting Network Communication page in Kyvos Manager, simplifying the setup process and enhancing security.

Querying Enhancements

• Static Function Support in Calculated Measures: Kyvos now supports the Static MDX function in calculated measures, providing faster query performance by evaluating expressions once and reusing them. This is particularly beneficial for improving the performance of large or complex expressions.

Cloud Support Enhancements

This release comes with the following enhancements for AWS, Azure, and GCP clouds.

• Autoscaling Support for Dedicated Compute Clusters: Kyvos now supports autoscaling of dedicated compute clusters. Instead of using fixed-size resources, the system automatically adjusts compute capacity based on data processing needs, improving performance, resource utilization, and cost efficiency.

• Web Portal and Kyvos Manager High Availability for Wizard-Based Deployments: Kyvos now supports Web Portal and Kyvos Manager High Availability for Kyvos wizard-based deployments, ensuring better reliability and uptime.

AWS Enhancements

• Support of Column-Level Encryption with RedShift: Kyvos now supports reading data from columns encrypted at the application level in Amazon Redshift using custom keys, enhancing column-level data security.

Azure Enhancements

• Automated Databricks Connector Setup for No-Spark Semantic Processing: This release automates the setup of Databricks connectors, storage credentials, and external locations for No-Spark semantic models, significantly reducing setup time and enhancing the user experience for processing No-Spark semantic models.

GCP Enhancements

• No-Spark Processing for Cross-Account BigQuery Data Sources: Kyvos now supports No-Spark processing for BigQuery data sources across different accounts.

Kyvos Manager Enhancements

This release comes with the following Kyvos Manager enhancements:

• Enhanced Node-Level Monitoring: Kyvos Manager now offers improved node-level monitoring, introducing new metrics under CPU, Memory, Disk, and Node Configuration categories on the Node Usage and Configurations page. These enhancements provide deeper visibility into system performance for better infrastructure management.
  Note: This enhancement is available across all supported cloud platforms and on-premises deployments.

Kyvos Viz. Enhancements

This release comes with the following Kyvos Visualization enhancements:

• New Pie Chart Customization Properties: Users can now personalize pie charts with features such as adjusting the outer radius, toggling label visibility, choosing label content (e.g., value, name, percentage), and rotating the chart. These enhancements provide more control and flexibility in data presentation.

• Metadata Support in CSV Export: Users can now export not only data values but also detailed metadata in CSV format. This includes sheet names, descriptions, semantic model information, field names, and applied filters, making the exported content more informative and useful.

• Export Data Cards as Images: Users can now export data cards from dashboards as image files. This makes it easier to share insights and include visuals in reports and presentations.

API Enhancements

This release comes with the following API enhancements:

Enhancements to Existing REST API:

• Access Token-based Authentication Support in Kyvos Rest APIs: Kyvos REST APIs now support access token-based authentication, enabling more secure and flexible integration.

• Execute Natural Language Query: The endpoint for executing natural language queries has been updated from /export/nlQuery to /dialogs/nlQuery for better alignment with Kyvos Dialogs.

New REST API

• Discovery of semantic model: The API returns a list of relevant semantic models based on a user's natural language query.

Behavior Changes

• Updated Default Instance Types for AWS and Azure: With this release, Kyvos has updated the default instance types as follows:

  • AWS: m6i.4xlarge

  • Azure: D4s_v5