What's new in IBM Software Hub

Version 5.3.1 of IBM Software Hub includes various fixes.

For details, see What's new and changed in the platform.

Run health checks for IBM Manta Data Lineage, Unstructured Data Integration, and Watsonx Code Assistant™ for Z Code Generation

You can now run service functionality checks for the following services by specifying the corresponding values when you run the cpd-cli health service-functionality command:

• IBM Manta Data Lineage: --services=datalineage
• Unstructured Data Integration: --services=udp
• Watsonx Code Assistant for Z Code Generation: --services=code-assistant-z-code-generation

For more information on running health checks, see the cpd-cli health service-functionality command reference.

Version 14.3.1 of the IBM Software Hub command-line interface includes various fixes.

For details, see What's new and changed in the IBM Software Hub command-line interface.

Access more data with new connectors

You can now set up connections to the following additional data sources:

• Amazon DynamoDB
• Google Drive
• Informatica PowerCenter
• Microsoft Sharepoint Files
• OpenSearch on IBM Cloud

Version 12.1.0 of the common core services includes various fixes.

For details, see What's new and changed in the common core services.

Version 1.61.0 of the scheduling service includes various fixes.

For details, see What's new and changed in the scheduling service.

Version 5.3.1 of the AI Factsheets service includes various fixes.

For details, see What's new and changed in AI Factsheets.

Version 5.3.1 of the Analytics Engine powered by Apache Spark service includes various fixes.

For details, see What's new and changed in Analytics Engine powered by Apache Spark.

Version 29.1.0 of the Cognos Analytics service includes various fixes.

For details, see What's new and changed in Cognos Analytics.

Version 9.1.0 of the Data Gate service includes various fixes.

For details, see What's new and changed in Data Gate.

Version 5.3.1 of the Data Privacy service includes various fixes.

For details, see What's new and changed in Data Privacy.

Version 5.3.1 of the Data Product Hub service includes various fixes.

For details, see What's new and changed in Data Product Hub.

Version 12.1.0 of the Data Refinery service includes various fixes.

For details, see What's new and changed in Data Refinery.

Version 5.3.1 of the Data Replication service includes various fixes.

For details, see What's new and changed in Data Replication.

Version 5.3.1 of the DataStage service includes various fixes.

For details, see What's new and changed in DataStage.

Audit user-initiated SQL statements to monitor user activity

Data Virtualization Admins can now audit SQL statements that users run to capture statement details including the time, username, and the status of the statement. Audited statements are then sent to the IBM Software Hub auditing system and then automatically forwarded to your configured SIEM solution.

See Auditing user-initiated SQL statements in Data Virtualization.

Version 3.3.1 of the Data Virtualization service includes various fixes.

For details, see What's new and changed in Data Virtualization.

Version 5.3.1 of the Db2 service includes various fixes.

For details, see What's new and changed in Db2.

Version 8.3.1 of the Db2 Big SQL service includes various fixes.

Version 5.3.1 of the Db2 Data Management Console service includes various fixes.

For details, see What's new and changed in Db2 Data Management Console.

Version 5.3.1 of the Db2 Warehouse service includes various fixes.

For details, see What's new and changed in Db2 Warehouse.

Version 12.1.0 of the Decision Optimization service includes various fixes.

For details, see What's new and changed in Decision Optimization.

This release of the EDB Postgres service includes various fixes.

For details, see What's new and changed in EDB Postgres.

Version 5.3.1 of the Execution Engine for Apache Hadoop service includes various fixes.

For details, see What's new and changed in Execution Engine for Apache Hadoop.

Manage relationships between governance artifacts and data quality assets on the governance rule level

You can now create and edit relationships between governance artifacts and data quality definitions or data quality rules on the governance rule level. Previously, you had to manage these relationships from the data quality definition asset or data quality rule asset view, which required edit permissions in production environments. Now, if you have the Manage data quality relationships permission, you can edit data quality rules and manage relationships.

Generate plain language descriptions for data quality rules

You can now automatically generate clear, plain language descriptions for your data quality rules, whether they are defined by business logic or written in SQL format. Plain language descriptions can help all users understand, review, and trust the data quality checks that are applied to your data assets.

Import, enrich, and assess the quality of data from additional data sources

You can now import metadata from the following data sources, enrich that data, and assess its quality:

• Amazon Aurora for MySQL
• Amazon Aurora for PostgreSQL

In addition, you can now write analysis output to tables in IBM Db2 for i.

Publish data quality rules to catalogs

You can now publish data quality rules from projects to catalogs. After a data quality rule is published into a catalog, you can add it into other projects for reuse.

Enhancements to roles and asset privacy settings for data source definitions

You now create data source definitions as public assets by default in the Platform assets catalog. This change improves visibility into newly created data source definitions.

• If you have the Viewer role in the Platform assets catalog, you can now see data source definitions on the Data source definitions tab.
• If you have the Editor or Admin role and have the necessary permissions and privileges, the improved visibility makes it easier to configure data lineage correctly.
• If you have the Viewer, Editor, or Owner role for a specific data source definition, you can now also see its endpoints.

Visualize asset owners and collaborators in Relationship explorer

You can now display relationships between assets and their owners and collaborators, which highly improves collaboration, clarifies ownership, and accelerates issue resolution or data onboarding.

Resynchronize data in Relationship explorer

Administrators can now resynchronize data to ensure that the latest updates are displayed in Relationship explorer. The resynchronization is useful to quickly correct out-of-sync data issues. Additionally, when Relationship explorer is enabled in environments where large volumes of assets and governance artifacts already exist, running resynchronization indexes all data, which then can be discovered.

Display data quality rules and data quality definitions in Relationship explorer

You can now visualize relationships for data quality rules and data quality definitions in Relationship explorer, including relationships with business terms and governance rules.

Make custom properties and relationships mandatory

When you create a custom property or relationship definition for governance artifacts, you can now decide whether to make those properties mandatory. When adding or editing an artifact with mandatory properties, the user must provide all mandatory values before saving.

Version 5.3.1 of the IBM Knowledge Catalog service includes various fixes.

For details, see What's new and changed in IBM Knowledge Catalog.

New data sources for lineage metadata import

You can now import lineage metadata from the following additional data sources:

• IBM Data Virtualization
• Informatica PowerCenter
• Microsoft SQL Server Analysis Services (SSAS)
• Microsoft Power BI Report Server (Microsoft Power BI Desktop)
• Statistical Analysis System (SAS)
• Talend

After the data is imported, you can visualize it on a lineage graph.

Connect to new data sources by using a new version of the Manta agent

You can now import lineage metadata from the following data sources by using an agent:

• Apache Hive
• Google BigQuery
• Informatica PowerCenter
• Microsoft SQL Server Integration Services (SSIS)
• Qlik Sense

Additionally, a new agent version is introduced, 1.4.0. Versions 1.0.0, 1.1.0, and 1.2.0 are deprecated. Consider updating your existing agent instances to the latest version 1.4.0.

You can now create assets from data sources that you use for lineage import

You can now create catalog assets by connecting to the following lineage-specific data sources:

• InfoSphere DataStage
• Informatica PowerCenter
• Microsoft Power BI (Azure)
• Microsoft SQL Server Analysis Services (SSAS)
• Microsoft SQL Server Integration Services (SSIS)
• MicroStrategy
• Qlik Sense
• Statistical Analysis System (SAS)
• Talend

Version 5.3.1 of the IBM Manta Data Lineage service includes various fixes.

For details, see What's new and changed in IBM Manta Data Lineage.

Compare master data records side by side

You can now compare two records in a split view to determine whether they belong together in an entity. Data stewards can explore and compare records and entities, view similarity analysis, and make informed decisions about entity composition.

Version 4.11.50 of the IBM Master Data Management service includes various fixes.

For details, see What's new and changed in IBM Master Data Management.

Read from IBM sources and write to Snowflake

When you deploy Data Collector engine version 7.0.0, you can read data from new sources and write data to a new target:

• Read data from IBM Cloud Object Storage
• Read data from IBM watsonx.data
• Stream data into Snowflake

For the complete list of new features in Data Collector 7.0.0, see 7.0.x Release Notes in the IBM StreamSets documentation.

Manage separate workspaces by using projects

You can now create projects to organize your work within Control Hub. Use projects as collaborative workspaces where you and your team can independently manage environments, pipelines, and jobs while your organization maintains centralized control.

For details, see Projects in the IBM StreamSets documentation.

Version 6.3.1 of the IBM StreamSets service includes various fixes.

For details, see What's new and changed in IBM StreamSets.

Version 10.1.0 of the Informix service includes various fixes.

Version 42.15.0 of the MANTA Automated Data Lineage service includes various fixes.

For details, see What's new and changed in MANTA Automated Data Lineage.

This release of the MongoDB service include various fixes.

For details, see What's new and changed in MongoDB.

Visually map relationships with the new GRC Canvas

Now you can use the GRC Canvas to visually map relationships between risks, controls, processes, or any other object. You can create and configure canvases for supported object types based on data that is already within OpenPages. You can add OpenPages objects or custom nodes and annotate with text nodes. Within the canvas, you can create and group nodes, and access your data directly from the canvas.

For more information, see Visualizing risk and compliance using Canvas.

Local MCP server for AI operations on objects

Now you can deploy a local MCP server to provide a standardized interface that enables AI agents to perform operations on OpenPages objects, such as issues and controls.

You can deploy a local MCP server within an agentic system, such as watsonx Orchestrate. This capability is provided as a beta feature with limited support in the 9.1.3 version of OpenPages. Try local MCP servers for test and development purposes only to explore agentic use cases. To deploy the MCP server on your local environment, use the open source IBM OpenPages local MCP server repository.

Connect with other AI models by using API extensions

Now you can connect OpenPages with any AI model, such as OpenAI ChatGPT, Mistral AI, other LLMs or ML endpoints. Administrators can create new integration plugins so that OpenPages deployments can virtually connect with any external solution that exposes an API.

Enhance API extensions with logging and traceability

Now the API extensions include logging and trace capabilities which improves serviceability and debugging for integration plugins.

Change settings to hide or view fields

You can now override visibility settings in Task Views and sections to make fields hidden or visible as needed.

Edit Microsoft Office files directly in OpenPages IBM Software Hub

You can now integrate Microsoft Office in OpenPages on IBM Software Hub. Open, edit, and save Microsoft Office files directly from OpenPages by using Microsoft Office applications that are installed on your Windows desktop.

Version 9.7.0 of the OpenPages service includes various fixes.

For details, see What's new and changed in OpenPages.

Reset pipeline caches from more places

To make it easier to reset pipeline job caches, you can now reset a cache from multiple locations:

• Pipelines view history
• Job details

When you reset the cache of a pipeline job run, all other job runs with the same name, or invocation ID, in that job are also reset.

Validate CEL expressions in real-time

You can now validate the Common Express Language (CEL) expressions in the Expression Builder as you create the expressions, which improves accuracy and reduces trial-and-error.

Improved navigation for nested pipeline jobs

You can now navigate between nested pipeline jobs and the view of each job's logs all within the Consolidated Logs view.

New concurrency setting for pipeline‑level job runs

You can now toggle the concurrency behavior for job runs that share the same name or invocation ID.

Version 5.3.1 of the Orchestration Pipelines service includes various fixes.

For details, see What's new and changed in Orchestration Pipelines.

Updated versions of Planning Analytics software

This release of the service provides the following software versions:

• Planning Analytics Workspace Version 3.1.4

  For details, see 3.1.4 - What's new in the Planning Analytics Workspace documentation.

• Planning Analytics Spreadsheet Services Version 3.1.4

  For details, see 3.1.4 - Feature updates in the TM1 Web documentation.

• Planning Analytics for Microsoft Excel Version 3.1.4

  For details, see 3.1.4 - Feature updates in the Planning Analytics for Microsoft Excel documentation.

• TM1 Database Version 12.5.6.

  For details, see 12.5.6 - What's new in the TM1 Database Version 12 documentation.

For more information about the software versions that are included in each release, see Component versions.

Version 5.3.1 of the Planning Analytics service includes various fixes.

Version 9.1.0 of the Product Master service includes various fixes.

For details, see What's new and changed in Product Master.

Version 12.1.0 of the RStudio Server Runtimes service includes various fixes.

For details, see What's new and changed in RStudio Server Runtimes.

Version 12.1.0 of the SPSS Modeler service includes various fixes.

For details, see What's new and changed in SPSS Modeler.

Version 12.1.0 of the Synthetic Data Generator service includes various fixes.

For details, see What's new and changed in Synthetic Data Generator.

Version 5.3.1 of the IBM Unstructured Data Integration service includes various fixes.

For details, see What's new and changed in IBM Unstructured Data Integration.

Version 1.13.0 of the Voice Gateway service includes various fixes.

Version 5.3.1 of the Watson Discovery service includes various fixes.

Version 5.3.1 of the Watson Machine Learning service includes various fixes.

For details, see What's new and changed in Watson Machine Learning.

Compute record-level metrics

You can now generate custom metrics that include record-level metrics.

Record-level metrics are metrics that are computed per row on the feedback or payload logging dataset, or on any other dataset that is used in a custom monitor.

For example, you might want to compute the HAP metric on user inputs and outputs for each record (for security or compliance tracking). You can do this by implementing a record-level metric. When users view the evaluation results, they can click a timestamp to see the user input and user output records.

For more information, see Creating custom evaluations and metrics.

Version 5.3.1 of the Watson OpenScale service includes various fixes.

New natural voices for Latin American Spanish and Australian English

You can now use the following additional natural voices with Watson Speech services:

• Latin American Spanish (es-LA) Alejandro Natural voice
• Latin American Spanish (es-LA) Daniela Natural voice
• Australian English (en-AU) Heidi Natural voice
• Australian English (en-AU) Jack Natural voice

For details, see Natural voices.

Version 5.3.1 of the Watson Speech to Text service includes various fixes.

For details, see What's new and changed in Watson Speech to Text.

Automatically stop notebook jobs when a cell fails

You can now control how notebook jobs behave when errors occur. When you configure a new notebook job, select Stop job run if an exception occurs to stop a notebook job as soon as a cell raises an exception. If the option is not selected, the job continues running even when errors occur.

Version 12.1.0 of the Watson Studio service includes various fixes.

For details, see What's new and changed in Watson Studio.

Version 12.1.0 of the Watson Studio Runtimes service includes various fixes.

For details, see What's new and changed in Watson Studio Runtimes.

New foundation models in watsonx.ai

You can now use the following foundation models for inferencing from the Prompt Lab and the API:

• ibm-defense-4-0-small
• devstral-small-2512
• devstral-medium-2512
• ministral-14b-instruct-2512
• mistral-large-2512

For details, see Foundation models in watsonx.ai.

Use text, audio, video, and images as context when you inference a custom foundation model

You can now add text, audio, video, and images as context when you inference any custom foundation model that has these capabilities.

For details, see Properties and parameters for custom foundation models.

Use audio input when you inference a foundation model with the watsonx.ai chat API

You can now add audio files encoded as binary data as context when you inference multimodal foundation models installed in your cluster that support audio input with the watsonx.ai chat API.

Pass model parameters when invoking deployed AutoAI RAG services

You can now include temperature and top_p model parameters when you invoke deployed AutoAI RAG services. With these additional settings, you have more flexible control over how the model responds for each request, instead of relying only on the default behavior. You can adjust the model’s tone and output without changing the deployment itself. 

New runtime requirement in AutoAI RAG experiments

AutoAI RAG now runs on the GenAI runtime, replacing the previous 24.1 runtime. The GenAI runtime is compatible with the latest foundation models, improves performance, and provides a more stable environment for RAG workloads. Make sure that you have the GenAI runtime installed in your environment to successfully run RAG experiments and deployments.

New model requirement for AutoAI RAG experiments with SQL knowledge bases

If you have any models that use SQL knowledge bases in AutoAI RAG, you must ensure that these models have the autoai_sql_rag function. AutoAI RAG uses this function to correctly evaluate and select models that work with SQL knowledge bases for retrieval.

Version 12.1.0 of the watsonx.ai service includes various fixes.

For details, see What's new and changed in watsonx.ai.

Version 5.3.1 of the watsonx Assistant service includes various security fixes.

Version 3.4.0 of the watsonx BI service includes various fixes.

Related documentation:

watsonx BI

Version 5.3.1 of the watsonx Code Assistant service includes various fixes.

For details, see What's new and changed in watsonx Code Assistant.

Related documentation:

watsonx Code Assistant

Version 5.3.1 of the watsonx Code Assistant for Red Hat Ansible Lightspeed service includes various fixes.

For details, see What's new and changed in watsonx Code Assistant for Red Hat Ansible Lightspeed.

Transform IMS programs that use CBLTDLI and EXECDLI interfaces

You can now use IMS Database Manager (DB) and IMS Transaction Manager (TM) calls to transform IMS programs that use the CBLTDLI and EXEC DLI interfaces to Java. However, you cannot transform IMS programs that use the EXEC DLI interface with multiple segments or IMS service calls.

Version 5.3.1 of the watsonx Code Assistant for Z service includes various fixes.

For details, see What's new and changed in watsonx Code Assistant for Z.

Version 2.7.1 of the watsonx Code Assistant for Z Agentic service includes various fixes.

Generate summaries for large PL/I files

You can now generate summaries for large PL/I files while continuing to use the Code Explanation feature for other code segments in the simple, detailed, or guided level of explanation.

Automatic expansion of copybooks and include files

IBM watsonx Code Assistant for Z Chat now automatically expands copybooks and INCLUDE files for COBOL, JCL, and PL/I programs.

Version 2.8.10 of the watsonx Code Assistant for Z Code Explanation service includes various fixes.

For details, see What's new and changed in watsonx Code Assistant for Z Code Explanation.

Version 2.8.10 of the watsonx Code Assistant for Z Code Generation service includes various fixes.

Use HLASM on Linux servers

You can now create, maintain, and debug applications on z/Architecture mainframes by using HLASM on Linux servers in addition to Windows servers.

Use configured agents to understand how containers interact with z/OS mainframe systems

The cli.sh script provides three new operations that you can use to test connectivity, scan PDS libraries, and download mainframe members directly from the command line.

Initiate Business Rule Discovery directly from the Agentic chat

You can use Agentic chat to initiate Business Rule Discovery in your application and describe them by using natural language.

Version 2.8.10 of the watsonx Code Assistant for Z Understand service includes various fixes.

For details, see What's new and changed in watsonx Code Assistant for Z Understand.

Version 2.3.1 of the watsonx.data service includes various fixes.

For details, see What's new and changed in watsonx.data.

Version 2.3.1 of the watsonx.data Premium service includes various fixes.

For details, see What's new and changed in watsonx.data Premium.

Export and import subflows with dependencies

You can now export subflows that include all of their dependencies. When you export a subflow, all required files are included in the exported output and restored on import. Subflows and their connections remain intact across projects. This enhancement applies to downloads, cpdctl exports, and Git sync operations.

Run multiple px-runtime replicas

You can now enable multiple px-runtime replicas for medium and large configuration sizes to improve scalability and resilience. Set the enableMultiPXReplicas flag to enable this feature. This flag is disabled by default.

Connect to Amazon Aurora for MySQL databases

You can now use the Amazon Aurora for MySQL connector in your DataStage flows. The connector includes full code generation and validation templates.

Read and write data from OpenSearch indices

You can now use the OpenSearch connector in your DataStage flows to read and write data from OpenSearch indices.

Read and write data from Amazon DynamoDB

You can now use the Amazon DynamoDB connector in your DataStage flows to read and write data from DynamoDB servers.

Read and write data from ClickHouse servers

You can now use the ClickHouse connector in your DataStage flows to read and write data from ClickHouse servers.

Monitor active StreamSets job runs

You can monitor the progress of an active StreamSets job run by reviewing the following information:

• Metrics: Include record counts, record throughput, and processing time for each stage in the job run.
• Log messages: Provide details about the job run.

Version 2.3.1 of the watsonx.data integration service includes various fixes.

For details, see What's new and changed in watsonx.data integration.

Add business context to document sets

You can now select an additional processing option for your unstructured data curation flows to enrich the generated document sets. If document content matches a data class in the selected scope, business terms and classifications that are linked to that data class are automatically assigned.

New data source for unstructured lineage import

You can now import unstructured lineage metadata from IBM Cloud Object Storage and visualize it on the lineage graph.

Visualize Unstructured Data Integration operators on the data lineage graph

When you visualize unstructured data on the lineage graph, you can now view individual operators within the flow. By reviewing runtime metadata and other details, you can better understand how each component contributes to the overall flow outcomes.

Manage relationships between governance artifacts and data quality assets on the governance rule level

You can now create and edit relationships between governance artifacts and data quality definitions or data quality rules on the governance rule level. Previously, you had to manage these relationships from the data quality definition asset or data quality rule asset view, which required edit permissions in production environments. Now, if you have the Manage data quality relationships permission, you can edit data quality rules and manage relationships.

Generate plain language descriptions for data quality rules

You can now automatically generate clear, plain language descriptions for your data quality rules, whether they are defined by business logic or written in SQL format. Plain language descriptions can help all users understand, review, and trust the data quality checks that are applied to your data assets.

Import, enrich, and assess the quality of data from additional data sources

You can now import metadata from the following data sources, enrich that data, and assess its quality:

• Amazon Aurora for MySQL
• Amazon Aurora for PostgreSQL

In addition, you can now write analysis output to tables in IBM Db2 for i.

Publish data quality rules to catalogs

You can now publish data quality rules from projects to catalogs. After a data quality rule is published into a catalog, you can add it into other projects for reuse.

Enhancements to roles and asset privacy settings for data source definitions

You now create data source definitions as public assets by default in the Platform assets catalog. This change improves visibility into newly created data source definitions.

• If you have the Viewer role in the Platform assets catalog, you can now see data source definitions on the Data source definitions tab.
• If you have the Editor or Admin role and have the necessary permissions and privileges, the improved visibility makes it easier to configure data lineage correctly.
• If you have the Viewer, Editor, or Owner role for a specific data source definition, you can now also see its endpoints.

Visualize asset owners and collaborators in Relationship explorer

You can now display relationships between assets and their owners and collaborators, which highly improves collaboration, clarifies ownership, and accelerates issue resolution or data onboarding.

Resynchronize data in Relationship explorer

Administrators can now resynchronize data to ensure that the latest updates are displayed in Relationship explorer. The resynchronization is useful to quickly correct out-of-sync data issues. Additionally, when Relationship explorer is enabled in environments where large volumes of assets and governance artifacts already exist, running resynchronization indexes all data, which then can be discovered.

Display data quality rules and data quality definitions in Relationship explorer

You can now visualize relationships for data quality rules and data quality definitions in Relationship explorer, including relationships with business terms and governance rules.

Make custom properties and relationships mandatory

When you create a custom property or relationship definition for governance artifacts, you can now decide whether to make those properties mandatory. When adding or editing an artifact with mandatory properties, the user must provide all mandatory values before saving.

New data sources for lineage metadata import

You can now import lineage metadata from the following additional data sources:

• IBM Data Virtualization
• Informatica PowerCenter
• Microsoft SQL Server Analysis Services (SSAS)
• Microsoft Power BI Report Server (Microsoft Power BI Desktop)
• Statistical Analysis System (SAS)
• Talend

After the data is imported, you can visualize it on a lineage graph.

Connect to new data sources by using a new version of the Manta agent

You can now import lineage metadata from the following data sources by using an agent:

• Apache Hive
• Google BigQuery
• Informatica PowerCenter
• Microsoft SQL Server Integration Services (SSIS)
• Qlik Sense

Additionally, a new agent version is introduced, 1.4.0. Versions 1.0.0, 1.1.0, and 1.2.0 are deprecated. Consider updating your existing agent instances to the latest version 1.4.0.

You can now create assets from data sources that you use for lineage import

You can now create catalog assets by connecting to the following lineage-specific data sources:

• InfoSphere DataStage
• Informatica PowerCenter
• Microsoft Power BI (Azure)
• Microsoft SQL Server Analysis Services (SSAS)
• Microsoft SQL Server Integration Services (SSIS)
• MicroStrategy
• Qlik Sense
• Statistical Analysis System (SAS)
• Talend

Version 2.3.1 of the watsonx.data intelligence service includes various fixes.

For details, see What's new and changed in watsonx.data intelligence.

Compute record-level metrics

You can now generate custom metrics that include record-level metrics.

Record-level metrics are metrics that are computed per row on the feedback or payload logging dataset, or on any other dataset that is used in a custom monitor.

For example, you might want to compute the HAP metric on user inputs and outputs for each record (for security or compliance tracking). You can do this by implementing a record-level metric. When users view the evaluation results, they can click a timestamp to see the user input and user output records.

For more information, see Creating custom evaluations and metrics.

Use assets with existing evaluations in Evaluation Studio

When you select the assets to compare in the Evaluation Studio, you can now select prompt templates and detached prompt templates that have already been evaluated. You no longer have to run evaluations for all assets to generate the comparison.

For more information, see Comparing assets with Evaluation Studio.

Create deployments for imported ML models in inventories by using the API or SDK

You can now use the API or SDK to create deployments for imported machine learning models in inventories. The deployment contains metadata only. The deployment is not an inference endpoint deployment of the model.

This feature provides both a representation for deployed models and a way to document evaluation metrics for models that are not served on watsonx.

For more information, see the API or SDK documentation.

Enhancements to Governance console

Questionnaire assessments for AI agents

The watsonx.governance Model Risk Governance solution includes the following changes to support the governance of AI agents:

• The new AI Agent Onboarding Risk Identification questionnaire assessment is used during the agent onboarding process to identify risks that are associated with the AI agent. The questions are independent of any specific use case and ask about the agent and its underlying model's training, inputs, and output behavior. This assessment is used in the AI Agent Onboarding workflow.
• The AI Use Case Risk Identification questionnaire assessment now includes a section for agentic risk identification. This new section can help you to identify potential risks that are associated an AI agent

Visually map relationships with the new GRC Canvas

You can use the new GRC Canvas to visually map relationships between risks, controls, processes, or any other object. You can create and configure canvases for supported object types based on data that is already within Governance console. You can add Governance console objects or custom nodes and annotate with text nodes. Within the canvas, you can create and group nodes, and access your data directly from the canvas.

Connect with other AI models by using API extensions

Now you can connect Governance console with any AI model, such as OpenAI ChatGPT, Mistral AI, other LLMs or ML endpoints. Administrators can create new integration plugins so that Governance console can virtually connect with any external solution that exposes an API.

For more information about new features and other updates in Governance console, see New features in version 9.1.3 in the OpenPages documentation.

Version 2.3.1 of the watsonx.governance service includes various fixes.

For details, see What's new and changed in watsonx.governance.

Use Slack to prompt for input for your flows and agents

Now, you can create flows or agents that prompt for input from Slack with form-based responses. Depending on your needs, these Slack responses can be Boolean, number, text, or single-choice fields. For setup instructions, see Enabling document-based interactions in chat.

Customize schemas for tools

You can customize input and output schemas for tools when schema editing is enabled. You can add, edit, or remove parameters and update properties such as name, description, and default values. By customising you can align tool schemas with your system requirements and validation rules. For details, see Customizing the schema for tools.

Inspect agentic workflow runs with flow inspector

You can use flow inspector to inspect an agentic workflow when the workflow is running. With the flow inspector, you can:

• View run status, initiator, agent, start time, environment, and instance ID.
• Search for specific runs, review flow events and parameters, and open the workflow in flow builder to update the configuration.
• Identify run patterns, trace execution paths, and validate workflow behavior across environments.

For details, see Inspecting agentic workflows.

Reuse your inputs in a chat

Agents can now reuse information that you provide during the same chat session. When an agent runs a workflow, such as collecting details through forms or completing an action, the data that you entered in a chat remains available for subsequent workflows in the same chat. With this improvement, you can continue the conversation without having to enter the provided information again.

Create and edit object data type properties by using the UI

You can now define and edit properties for an object data type by using the user interface instead of manually writing JSON schemas. When you configure object data types for inputs, outputs, flow variables, or generative prompts, you can visually define the object structure, properties, and default values. You can switch between a simple view and a JSON view as needed. For details, see Adding properties to an object data type.

Import external A2A agents as collaborators

You can now add external agents that use the Agent to Agent (A2A) protocol as collaborators when you build an agent in the UI. You are easier to integrate third-party systems that follow the A2A standard and control how they exchange data during execution. For details, see Adding agents from third-party platforms.

Add a phone-like preview for testing DTMF in voice agents

You can now use phone-like preview in voice chats to test Dual-Tone Multi-Frequency (DTMF) settings without ending the voice chat or making an actual phone call. You can use the keypad to send DTMF inputs (0–9, *, #) to the agent and validate DTMF-dependent flows. You can also turn off the setting without ending the ongoing voice session. For details, see Configuring voice settings for agents.

Add DTMF support for Genesys audio connector

You can now enable Dual-Tone Multi-Frequency (DTMF) input for phone calls that use Genesys audio connector. By using DTMF, your voice agents can collect keypad inputs during a call, IVR-style flows, and mixed voice–keypad interactions. You can configure DTMF settings in Manage > Voice when you create or edit a voice configuration. For details, see Configuring voice settings for agents.

Add Deepgram as a Speech to Text and Text to Speech provider

You can now select Deepgram as a Speech to Text (STT) and Text to Speech (TTS) provider when you create or edit a voice configuration gives you more language options and faster response times for voice interactions. To use Deepgram, go to Manage > Voice, create or edit a voice configuration, and choose Deepgram in the STT or TTS provider list. For details, see Configuring voice settings for agents.

Connect ready configured Apps with Pre-Built Tools and Agents

watsonx Orchestrate detects authentication requirements for pre-built tools and agents automatically. The system reads security requirements from tool metadata and provides step-by-step configuration guidance. You can bind app connections without manual configuration, reduce configuration errors and save time on onboarding

Pro-code extensions for Agent control

Pro-code extensions for Agent control gives you or your organizations to implement custom policies, security controls, and guardrails for watsonx Orchestrate agents through programmable hooks that intercept agent operations before and after execution. The feature helps the hook to run before the chain-of-thoughts is built, hook to control direct agent access, and hook runs after agent generates final response.

Integrate with custom header names

You can connect to backend systems that use custom API key header names such as api-key, x-apikey, or myapikey. The system reads the API key name and location directly from tool definitions, removing the limitation where authentication worked with fields named apikey.

Control how you use streaming responses

You can scroll through long agent responses without forced auto-scroll. Use the pause auto-scroll toggle to lock your viewport while generation continues, then select jump to latest to resume live updates instantly.

Copy agent responses instantly

You can copy any agent response to your clipboard with one click. Eliminate manual text selection and highlighting for long content blocks. Now, you can reuse the content, share, or store generated content efficiently.

Manage notifications, tasks across chat threads

You can view the pending activities across all chat threads in a unified activity queue. Work between prompts, async jobs, tempus tasks, and workflows initiated by other users or the system. Use clear badges and notifications for cross-thread updates, and select items from an intuitive picker.

Control monitoring and token consumption for agent

You can now choose which agents to monitor instead of processing messages across all agents. Enable or disable monitoring at the agent level, view token consumption per agent, and set monitoring priorities based on importance.

Expand agent capabilities with remote tools (MCP servers)

You can connect to remote MCP (Model Context Protocol) servers securely from the UI. Preview available tools, add them to your agents, and manage tools without reconfiguring connections. Refresh tool lists from remote servers and reset connections as needed.

Share published agents across your organization

You can share and reuse agents and related artifacts from tenant-wide from workspace. Peers can copy assets with one click, preserving intent and metadata in Agent card.

Version 7.1.0 of the watsonx Orchestrate service includes various fixes.

For details, see What's new and changed in watsonx Orchestrate.

If you have an existing IBM Instana Observability installation, you can deploy an Instana agent on the cluster where IBM Software Hub is installed so that you can send data to IBM Instana Observability.

You can enable Instana metrics collection for the platform and services to see real-time data and root-cause analysis so that you can quickly act on performance and stability issues that affect IBM Software Hub.

For details, see Integrating IBM Software Hub with IBM Instana Observability.

If you have multiple instances of IBM Software Hub, you can use a web-based Git repository to manage assets across instances. For example, you can use GitHub to promote assets from your development instance to your staging instance to ensure that the assets are the same on both environments.

You can use the following web-based Git repositories with IBM Software Hub:

• GitHub
• GitHub Enterprise
• GitLab
• GitLab Self-Managed

For details, see:

• Enabling integration Git for asset promotion
• Managing asset promotion with Git

Previously, if you wanted to enable automatic scaling through the Horizontal Pod Autoscaler, you needed to use the cpd-cli. You can now enable and disable automatic scaling from the IBM Software Hub web client.

Access more data with new connectors

• Amazon Aurora for MySQL
• Amazon Aurora for PostgreSQL
• ClickHouse
• Confluence
• Datastax HCD
• Iceberg Metastore

Write data to Microsoft Azure Data Lake Storage with the watsonx.data connection

You can now write data to the Microsoft Azure Data Lake Storage data source by using the watsonx.data connection.

Write data that uses the compressed gzip format for Amazon S3

You can now write in the compressed data format gzip for the Amazon S3 data source.

You can load data that is in this format into the Snowflake connection.

Access new versions of your data sources

You can now connect to updated versions of the following data sources to take advantage of the latest features and improvements:

• PostgreSQL version 18

Version 12.0.0 of the common core services includes various fixes.

For details, see What's new and changed in the common core services.

Schedule ARM CPUs and GPUs on remote physical locations

Premium If you use remote physical locations to expand your IBM Software Hub deployment to remote clusters, you can now use the scheduling service to schedule ARM CPUs and ARM GPUs on the remote physical locations.

If the remote cluster has a cluster autoscaler, you can use the following options:

• Use the --max_cpu_arm option to allow the scheduling service to schedule additional ARM-based CPUs if the workload exceeds the current available ARM CPU.
• Use the --max_gpu_arm option to allow the scheduling service to schedule additional ARM-based GPUs if the workload exceeds the current available ARM GPU.

For more information, see Registering a remote physical location with an instance of IBM Software Hub.

Restriction: You cannot schedule ARM-based CPU or ARM-based GPU on existing remote physical locations. If you want to schedule GPUs, you must either:

• Create a new remote physical location
• Re-create the physical location by running the following commands:
  1. cpd-cli manage delete-physical-location
  2. cpd-cli manage create-physical-location
  3. cpd-cli manage register-physical-location

Version 1.60.0 of the scheduling service includes various fixes.

For details, see What's new and changed in the scheduling service.

Version 5.3.0 of the AI Factsheets service includes various fixes.

For details, see What's new and changed in AI Factsheets.

Version 5.3.0 of the Analytics Engine powered by Apache Spark service includes various fixes.

Manage fonts and style sheets with the Artifacts API

The new version, 2.3.0, of the Cognos Analytics Artifacts API contains new artifacts: fonts and stylesheets .Now you can use the API to list, upload, download and delete font and stylesheet files. For details, see Managing artifacts with Cognos Analytics APIs.

Updated software version for Cognos Analytics

This release of the service provides Version 12.1.1 of the Cognos Analytics software. For details, see Release 12.1.1 in the Cognos Analytics documentation.

Version 29.0.0 of the Cognos Analytics service includes various fixes.

For details, see What's new and changed in Cognos Analytics.

Updated software version

This release of the service provides Version 12.1.1 of the Cognos Analytics dashboards software. For details, see Release 12.1.1 - Dashboards in the Cognos Analytics documentation.

Version 5.3.0 of the Cognos Dashboards service includes various fixes.

For details, see What's new and changed in Cognos Dashboards.

Remote Db2 support for Power

You can now connect an additional type of target database to Data Gate: a remote Db2 target database that is running on a Power computer (PPC64LE architecture). For details, see Connecting to a remote Db2 instance.

Certificates signed by third-party CA

Data Gate now supports SSL-encrypted connections to the Db2 or Db2 Warehouse target database that are secured with certificates signed by an external certificate authority (CA).

Version 9.0.0 of the Data Gate service includes various fixes.

For details, see What's new and changed in Data Gate.

Duplicate data protection rules

You can now duplicate existing data protection rules or edit the rule details to create new rules.

Activate and deactivate data protection rules in the UI

You no longer have to delete rules that you don't want to use, and then recreate them later if you need them again. Now, you can deactivate and activate data protection rules and revoke the rules without deleting them. See and manage the status of any rule on the Rules page.

Version 5.3.0 of the Data Privacy service includes various fixes.

For details, see What's new and changed in Data Privacy.

Create and access data contracts in Open Data Contract Standard v3

Streamline your management of data contracts by using Open Data Contract Standard v3 (ODCS v3) format in Data Product Hub

• Producers: You can now create data contracts in ODCS v3 format. Create contracts from scratch or by using a predefined template.

• Consumers: You can access and review data contracts directly in Data Product Hub or download them in YAML format, along with any associated test status information.

Deliver data products from Azure Databricks

You can now subscribe to a data product that is created in Azure Databricks by using the Access in Azure Databricks delivery method. Consumers can directly access Azure Databricks resources by using Data Product Hub. After delivery of the data products, consumers see details on how to access the specific resources in Azure Databricks.

Deliver data assets to a project by using the access in watsonx.data delivery method

You can now choose to import data product assets to a project by using the access in watsonx.data delivery method.

Manage and view data product reviews

Consumers can now create, edit, and delete reviews of data products. Producers cannot manage reviews.

Version 5.3.0 of the Data Product Hub service includes various fixes.

For details, see What's new and changed in Data Product Hub.

Version 12.0.0 of the Data Refinery service includes various fixes.

For details, see What's new and changed in Data Refinery.

Replicate range-partitioned Db2 tables to supported target data stores

You can now use the Data Replication service to replicate Db2 tables that are partitioned based on the range of values in one or more columns. These types of tables are also known as range-partitioned tables. You can replicate range-partitioned tables to Db2 on Cloud, Db2 Warehouse on Cloud, IBM watsonx.data, and Apache Kafka target data stores.

Version 5.3.0 of the Data Replication service includes various fixes.

For details, see What's new and changed in Data Replication.

Connect to Microsoft Fabric Warehouse

You can now connect to Microsoft Fabric Warehouse using the new connector in DataStage, enabling seamless integration with your data workflows.

Use IBM Instana to validate your workload runs

You can now automatically discover, monitor and visualize components of your cluster in real time. With IBM Instana observability features, you can efficiently detect and address performance problems, minimizing the time spent on troubleshooting

Version 5.3.0 of the DataStage service includes various fixes.

For details, see What's new and changed in DataStage.

Enable high concurrency and greater scalability of query processing by using Data Virtualization agents

Data Virtualization agents now run in their own dedicated pods instead of within the Data Virtualization primary head pod, for better system scalability.

• For new installs, the number of Data Virtualization agent pods are automatically provisioned based on the sizing option you choose in the web client.

• If your existing Data Virtualization instance uses custom sizing, then upgrading your Data Virtualization instance automatically adds five agent pods, each requiring two CPUs. The increased resource usage is typically balanced if your custom cluster was deployed with sufficient resources to accommodate the extra load without dropping below a stable minimum. However, if you have custom sizing and limited resources, then you might experience a net increase in resource usage.

  To customize the number of Data Virtualization agent pods, or adjust the CPU usage and memory settings, see Customizing the pod size and resource usage of Data Virtualization agents.

Automatically apply personal credentials setting when importing data sources

Personal credentials are now enabled by default in Data Virtualization. When you add a data source on the platform connection with personal credentials turned on, the same setting is automatically applied when the data source is imported into Data Virtualization.

In order to successfully access, virtualize, and query the data source through Data Virtualization, each user needs to configure their own credentials on the platform connections side.

Migrate Data Virtualization assets to and from your Git repository

You can now export and import your Data Virtualization assets across different environments (for example, from development to QA or production) from your Git repository by using Data Virtualization APIs. By using Git, Admin users can quickly synchronize assets, like tables, nicknames, and views, by promoting the Data Virtualization objects from a Data Virtualization instance to a Git branch, and by pulling updates from Git back into Data Virtualization.

You can migrate the following objects with Git:

• Nicknames (excluding those with personal credentials)
• Schemas
• Tables (excluding those with personal credentials)
• Views
• Authorization statements (GRANTs)
• Statistics

See Migrating Data Virtualization objects by using Git.

Use new views to simplify troubleshooting and admin tasks

• You can now troubleshoot connection failures by using automated diagnostic tests. When a datasource connection fails, Data Virtualization automatically runs a series of connectivity tests (including Ping, OpenSSL, Netcat, and Traceroute) to identify the root cause. The results are logged in ConnectivityTest.log on each qpagent, along with a unique DIAGID included in the error message which you can use with the LISTCONNECTIVITYTESTWARNINGS view to retrieve detailed logs. The DIAGID is cleared when the datasource connection becomes available again.

• You can now display the list of columns of the tables in a RDBMS source by using the LISTCOLUMNS view.

• You can now set configuration properties specific to Data Virtualization and Federation directly for your Data Virtualization connection by using the SETCONNECTIONCONFIGPROPERTY and SETCONNECTIONCONFIGPROPERTIES stored procedures. Additionally, you can now set Federation-specific options for existing SETRDBCX procedures.

• See the full list of Stored procedures and Views.

Grant collaborators the new INSPECT data source privilege to view source metadata

You can now grant INSPECT privilege to users or the DV_METADATA_READER role to enable those users import lineage metadata with MANTA Automated Data Lineage.

To get started in the web client, navigate to the Data sources page and then select the Manage access setting on your data source, and then select the grantees. You can also grant the INSPECT privilege to the DV_METADATA_READER role by selecting Grant INSPECT privilege to the DV_METADATA_READER role. In the INSPECT column, you can grant or revoke the INSPECT privilege for the grantee.

See INSPECT privilege in Data source connection access restrictions in Data Virtualization and Configuring Data Virtualization connections for lineage imports.

Connect to Apache Cassandra and watsonx.data Presto data sources

You can now connect to Apache Cassandra and watsonx.data Presto from Data Virtualization.

Version 3.3.0 of the Data Virtualization service includes various fixes.

For details, see What's new and changed in Data Virtualization.

Deploy Db2 database with non-root deployment (Restricted-v2 SCC)

Now, when deploying a new Db2 database in your IBM Software Hub cluster, you can enable non-root deployment by selecting the checkbox Deploy Db2 with non-root deployment on the Advanced Configurations page. Selecting the Restricted-v2 option uses Red Hat OpenShift’s default restricted-v2 Security Context Constraint (SCC) to meet strict security requirements while maintaining full functionality.

This SCC ensures that:

• Workloads run with non-root privileges.
• Use of sudo or elevated permissions is not allowed.

For more information on permission levels and requirements, see Deploying Db2 with non-root access in a restricted-v2 SCC on IBM Software Hub.

Version 5.3.0 of the Db2 service includes various fixes.

For details, see What's new and changed in Db2.

Version 8.3.0 of the Db2 Big SQL service includes various fixes.

For details, see What's new and changed in Db2 Big SQL.

Version 5.3.0 of the Db2 Data Management Console service includes various fixes.

For details, see What's new and changed in Db2 Data Management Console.

Deploy Db2 Warehouse database with non-root deployment (Restricted-v2 SCC)

Now, when deploying a new Db2 Warehouse database in your IBM Software Hub cluster, you can enable non-root deployment by selecting the checkbox Deploy Db2 Warehouse with non-root deployment on the Advanced Configurations page. Selecting the Restricted-v2 option uses Red Hat OpenShift’s default restricted-v2 Security Context Constraint (SCC) to meet strict security requirements while maintaining full functionality.

This SCC ensures that:

• Workloads run with non-root privileges.
• Use of sudo or elevated permissions is not allowed.

For more information on permission levels and requirements, see Deploying Db2 Warehouse with non-root access in a restricted-v2 SCC on IBM Software Hub.

Query external data with Datalake tables

You can now use Datalake tables to work with data stored in open formats like PARQUET and ORC directly from Db2 Warehouse, without moving the data into the database.

With Datalake tables, you can do the following:

• Query external data: Define a Datalake table in Db2 Warehouse and use it in complex queries with other Db2 Warehouse tables.
• Export Db2 data to object storage while keeping it queryable by using:
  • INSERT
  • SELECT INFO
  • CREATE DATALAKE TABLE AS SELECT
• Import data from a Datalake table into a table in the database. You can perform operations such as casts, joins, and dropping columns to manipulate data during importing.

Version 5.3.0 of the Db2 Warehouse service includes various fixes.

For details, see What's new and changed in Db2 Warehouse.

Compare models in Decision Optimization experiments

You can now compare models from different scenarios in a Decision Optimization experiment and compare the log files when the models are solved. When you compare models this way, you can see the different scenarios side by side.

Version 12.0.0 of the Decision Optimization service includes various fixes and updates.

For details, see What's new and changed in Decision Optimization.

This release of the EDB Postgres service includes various fixes.

For details, see What's new and changed in EDB Postgres.

Version 5.3.0 of the Execution Engine for Apache Hadoop service includes various fixes.

Create SQL-based assets and data quality rules with text instead of SQL

Now you can describe the data asset or the data quality rule that you want to create in plain English and convert this text query into an SQL query. You can then run the generated query to create the asset or the rule.

Tech preview This is a technology preview and is not supported for use in production environments.

Disable certain generative AI capabilities for selected projects

Even if the product is installed with generative AI capabilities, you might not want to use these capabilities in all of your projects. You can now disable these capabilities per project. In projects where the capabilities are disabled, you can't work with natural language queries to create SQL-based assets and data quality rules. In addition, LLM-based name, description, or term generation and term assignment in metadata enrichment are disabled.

Define catalog-specific custom properties for assets

You can now restrict custom properties for assets to a specific catalog. By using catalog-specific custom properties, you can more effectively display values that pertain only to selected domains and ensure that the right information is available to the right users.

To list custom properties that are restricted by a given catalog, use the sort by scope option and scroll down to the items for the catalog that you're interested in.

Manage columns for catalogs

You can now select which columns to display in the asset listing grid by clicking Manage columns on the catalog page. Select your columns, reorder them if necessary, and save your preferences to keep the information that is most relevant for you readily available. For example, you can modify the view to show you a list of assets with the display name, owners, and date added columns only.

Optimize term assignment

With the new tuning options for term assignment, now you can influence the weighting of term suggestions for better precision or recall.

Import primary keys and foreign keys and visualize them in Relationship Explorer

Import primary keys and foreign keys with metadata import instead of metadata enrichment. After import, you can access the associated relationships through the RHS panel and Relationship Explorer.

Versioning of governance artifacts

Track historical changes for the artifacts, schedule new versions to be published in the future, and restore or archive previous versions with the new Versions panel.

Version 5.3.0 of the IBM Knowledge Catalog service includes various fixes.

For details, see What's new and changed in IBM Knowledge Catalog.

Export data lineage to Collibra

You can now export data lineage and view it in Collibra. If you transfer lineage information into Collibra data governance platform, you can see a comprehensive view of your data flows and dependencies within your governance framework.

Starting parents are introduced in the data lineage graph

When you select an asset to be a starting asset in the lineage, all assets that are higher in the hierarchy are marked as starting parents. Also, all child assets of the selected asset are marked as starting assets. This distinction clarifies which assets are selected as the starting points for the lineage.

Version 5.3.0 of the IBM Manta Data Lineage service includes various fixes.

For details, see What's new and changed in IBM Manta Data Lineage.

IBM Match 360 is now known as IBM Master Data Management

The IBM Match 360 service is renamed to IBM Master Data Management.

View historical data for entities, records, and relationships in your master data

You can now view the history of each entity, record, and relationship in your master data and compare historical attribute values to the current version. Select any past update to view the attribute values at that point in time and also see whether each update was initiated by a user, source system, or linkage action. You can use this capability to help with audit tracking and analysis of data changes over time.

A data engineer can configure whether the service keeps historical data. Storing history details increases the storage requirements of your database.

Configure potential match workflows for each entity type

You can now configure a potential match workflow for each entity type in your data model. Potential match workflows identify matching issues within your data, then create and assign tasks for data stewards to resolve them. Potential matches are now also searchable by source to help data stewards to identify and resolve potential duplicate records and entities. Configure all your task workflows from the Task types page.

Edit any attribute of a master data entity, including composite values

You can now edit any of an entity's attribute values, even if the original value was derived from the entity's member records by applying attribute composition rules. After a data steward overrides a composite value, the user-defined value will continue to be maintained even if the composition of the entity changes.

This capability is available only for entity types that a data engineer has configured to enable entity persistence.

View more information about group types and hierarchy types

When you open a hierarchy type or group type from the data types pages, you can now see more details at a glance, including who created the type, when they created it, and a list of group or hierarchy instances that are based on the selected type. You can also navigate directly from the data types page to the Workspace view to manage each hierarchy and group instance, and its members.

Version 4.10.28 of the IBM Master Data Management service includes various fixes.

For details, see What's new and changed in IBM Master Data Management.

Version 6.3.0 of the IBM StreamSets service includes various fixes.

For details, see What's new and changed in IBM StreamSets.

Version 10.0.0 of the Informix service includes various fixes.

For details, see Fix list for Informix Server 15.0.0.2 release.

Version 42.14.1 of the MANTA Automated Data Lineage service includes various fixes.

For details, see What's new and changed in MANTA Automated Data Lineage.

This release of the MongoDB service include various fixes.

For details, see What's new and changed in MongoDB.

Version 9.6.0 of the OpenPages service includes various fixes.

See the parameters of utility scripts in the Expression Builder

When you use the Expression Builder, you can now see the parameters (inputs) that the utility scripts require. This enhancement shows you what arguments you need to input, which leads to fewer errors.

The utility scripts run using a CEL (Common Expression Language) expression.

New default embedded runtime

You can now use a new default embedded runtime, removing the dependency on the Tekton runtime. This update eliminates compatibility issues with OpenShift Pipelines, ensuring stable co-existence. The new runtime delivers faster performance and greater scalability for your pipeline workloads.

Version 5.3.0 of the Orchestration Pipelines service includes various fixes.

For details, see What's new and changed in Orchestration Pipelines.

Updated versions of Planning Analytics software

This release of the service provides the following software versions:

• Planning Analytics Workspace Version 3.1.2

  For details, see 3.1.2 - What's new in the Planning Analytics Workspace documentation.

• Planning Analytics Spreadsheet Services Version 3.1.2

  For details, see 3.1.2 - Feature updates in the TM1 Web documentation.

• Planning Analytics for Microsoft Excel Version 3.1.2

  For details, see 3.1.2 - Feature updates in the Planning Analytics for Microsoft Excel documentation.

For more information about the software versions that are included in each release, see Component versions.

Version 5.3.0 of the Planning Analytics service includes various fixes.

Categorize multiple items at once

You can now categorize more than one item at a time by using the bulk categorization capability on the Data explorer, Search, or Free text search pages.

Create rules on attribute collections 

You can now create rules on attribute collections, and across different specifications.

Generate report with only selected data

In the Generate report feature, you can now specify to include only the selected attributes, rows, and columns that get included in the exported Microsoft Excel worksheet.

Suspect Duplicate Processing with Data survivorship 

Now, when you run Suspect Duplicate Processing, the Data survivorship rules improve how IBM Product Master matches and merges duplicate records. You can also match data that is still being processed.

Enhanced Data completeness dashboard

You can now use the Data completeness dashboard to check the completeness of your data based on categories and items within workflows.

Dashboard builder enhancements

You can now use the new Criteria console on the Dashboards > Charts tab to view the mapping of all criteria to their associated charts.

You can now drag and drop charts, click select chart data to view details, and also export dashboard data from the Dashboard tab.

Version 9.0.0 of the Product Master service includes various fixes.

For details, see What's new and changed in Product Master.

Version 12.0.0 of the RStudio Server Runtimes service includes various fixes.

Override storage type in a Data Asset node

You can now easily override the inferences that the Data Asset node makes about storage types when it imports data. The Data Asset node reads a sample of the data that it imports to infer what type of data is in a table field, such as an integer, date, or string. Previously, changing the storage type after this inference was difficult. Now you can override the storage type for a field in the settings for the Data Asset node.

Analyze Chinese text data in SPSS Modeler with Text Analytics

You can now use the Text Analytics nodes in SPSS Modeler to analyze text data that is written in Simplified Chinese. Text Analytics nodes use advanced linguistic technologies and text mining techniques to analyze text data and extract concepts, patterns, and categories.

Save visualizations to the project

When you create an SPSS Modeler job, you can now choose to save any visualizations that are generated to the project. For example, you can save the graphical outputs from a Chart node, Plot node, or Evaluation node. You can view the files in the Assets tab and download them.

Version 12.0.0 of the SPSS Modeler service includes various fixes.

For details, see What's new and changed in SPSS Modeler.

Create jobs for generating unstructured synthetic data by using the user interface

In addition to using the REST API, you can now use the user interface to create jobs for generating unstructured synthetic data.

Use multi-table nodes to generate synthetic data with referential integrity

You can now create synthetic data by using several data tables from a database connection instead of just one data table. The multi-table nodes use referential integrity to preserve parent–child dependencies across multiple tables, and they also preserve distributions and correlations within the same table. When you generate synthetic data by using these nodes, the synthetic data mimics all these dependencies, which allows you to make synthetic data that more closely mimics the entire dataset.

Scripting with Synthetic Data Generator

You can now programmatically make changes to nodes and flows in Synthetic Data Generator by using scripts written in Python or Jython. Use either the Scripting pane or embedded API to run your scripts. You can use scripting to automate tasks that are highly repetitive or to quickly change all of the nodes in a flow.

Scripting with parameters

You can also use parameters in the Python scripts that you write for Synthetic Data Generator. You can use parameters as a way of passing values at runtime rather than hard coding them directly in a script. This flexibility makes it easier to adjust settings for a Synthetic Data Generator flow while reusing the same scripts.

Override storage type in an Import node

You can now override the inferences that the Import node makes about storage types when it imports data. The Import node reads a sample of the data that it imports to infer what type of data is in a table field, such as an integer, date, or string. Previously, changing the storage type after this inference was difficult. Now you can easily override the storage type for a field in the settings for the Import node.

Save evaluation metrics to the project

When you create an Synthetic Data Generator job, you can now choose to save the data that is generated when an Evaluate node runs. You can then view the data in the Assets tab and download it.

Version 12.0.0 of the Synthetic Data Generator service includes various fixes.

For details, see What's new and changed in Synthetic Data Generator.

Version 1.12.0 of the Voice Gateway service includes various fixes.

Version 5.3.0 of the Watson Discovery service includes various fixes.

For details, see What's new and changed in Watson Discovery.

Deploy your Python code, R code, and Tensorflow models by using the latest software specifications

For R scripts, you can now use the latest software specification that is based on R 4.4 to deploy your code.

You can also deploy Python functions, Python scripts, and Tensorflow models with the new runtime-25.1-py3.12-cuda version of the 25.1 software specification and use GPUs to inference your code.

Optimize AutoAI for Machine Learning experiments across hyperparameter tuning for score and runtime

You can now optimize hyperparameter tuning for the highest score and shortest runtime when you run AutoAI for Machine Learning experiments. Previously, this optimization was available only during algorithm selection. Now, you can apply score and runtime optimization during hyperparameter tuning or across both phases for a balanced approach, helping you build high-performing models faster while reducing training time.

Version 5.3.0 of the Watson Machine Learning service includes various fixes.

For details, see What's new and changed in Watson Machine Learning.

Version 5.3.0 of the Watson OpenScale service includes various fixes.

Spoken language identification

Language Identification (LID) now automatically detects spoken languages in audio streams. The model continuously analyzes incoming audio and returns the identified language when it reaches a certain confidence threshold. The model updates the detected language throughout the session and enables early session termination after the initial response. The client applications can switch languages for downstream processing, making it ideal for multilingual environments such as call centers. For details, see Spoken language identification.

Speech transcript enrichment

Speech transcript enrichment improves the readability and usability of raw Automatic Speech Recognition (ASR) transcripts. This post-processing service adds punctuation and applies intelligent capitalization to enhance the structure and clarity of spoken content. For details, see Speech transcript enrichment. For details on how to configure the enrichment features, see Configure Speech transcript enrichment.

Natural voices

You can now use the following natural voices with Watson Speech services:

• US English (en-US) Ellie Natural voice
• US English (en-US) Emma Natural voice
• US English (en-US) Ethan Natural voice
• US English (en-US) Jackson Natural voice
• US English (en-US) Victoria Natural voice
• UK English (en-GB) Chloe Natural voice
• UK English (en-GB) George Natural voice
• CA English (en-CA) Hannah Natural voice
• Brazilian Portuguese (pt-BR) Lucas Natural voice
• Brazilian Portuguese (pt-BR) Camila Natural voice

For details, see Natural voices.

Version 5.3.0 of the Watson Speech to Text service includes various fixes.

For details, see What's new and changed in Watson Speech to Text.

Secure project by using permanent admins

You can now set yourself or other users as permanent admins to keep projects secure and prevent orphaned projects. If you have permission to manage projects, you become a permanent admin automatically when you create or join a project. Permanent admins have more permissions than regular admins. They can maintain full control of projects and can manage other permanent admins.

Easily maintain clear project documentation for projects with deprecated Git integration

You can now create and manage project documentation directly in the new Documentation editor for projects that have deprecated Git integration. By using the Documentation editor, you can easily organize details, collaborate with your team, and maintain clear records. Access the Documentation editor from the project Overview page.

Version 12.0.0 of the Watson Studio service includes various fixes.

For details, see What's new and changed in Watson Studio.

Run Python and R code by using environments that are based on Spark 4 engine

You can now run your Python and R code by using the latest environments that are based on Spark 4 engine.

Run R 4.4 code in Jupyter Notebooks

You can now Run R 4.4 code in Jupyter Notebooks by using Runtime 25.1 on R 4.4

Version 12.0.0 of the Watson Studio Runtimes service includes various fixes.

For details, see What's new and changed in Watson Studio Runtimes.

New foundation models in watsonx.ai

You can now use the following foundation models for inferencing from the Prompt Lab and the API:

• granite-4-h-tiny
• granite-docling-258M
• ibm-defense-4-0-micro

For details, see Foundation models in watsonx.ai.

Access models across multiple providers with the model gateway

You can now securely configure and interact with foundation models from multiple providers with the model gateway by using the API. In addition, you can manage the model gateway with integrated load-balancing, access policies, and rate limits.

Capture semantic meaning and refine retrieved results by using custom embedding and reranking models

You can now add custom embedding and reranking models to watsonx.ai and use them to capture semantic meaning and refine retrieved results.

For details, see Registering custom foundation models for global deployment.

All custom foundation models now use the vLLM inferencing server

All custom foundation models now use the vLLM inferencing server. If your deployed models use the TGIS inferencing server, you might have to migrate them.

New text classification API in watsonx.ai

You can now use the new text classification method in the watsonx.ai REST API to classify your document before you extract textual content to use in a RAG solution.

You can classify your document with the classification API into one of several supported common document types without running a longer extraction task. By pre-processing the document, you can then customize your text extraction request to efficiently extract relevant details from your document.

New vector index transactional API in watsonx.ai

You can now use the new vector index transactional API methods to create and manage vector index assets in a project.

Use the user interface for Synthetic Data Generator to generate unstructured synthetic data

The user interface for creating jobs to generate unstructured synthetic data is now generally available. The user interface in Synthetic Data Generator makes creating and running jobs easier by organizing all the settings and seed document requirements into simple options and fields.

Improvements for AutoAI for RAG experiments

You can now use the following features for your AutoAI for RAG experiments:

• Use semantic chunking in AutoAI for RAG experiments

  You can now use the semantic chunking method to break down documents in an AutoAI for RAG experiment. Semantic chunking splits documents based on meaning, making it well-suited for complex or unstructured data.

• Use chat API models in AutoAI for RAG experiments

  You can now use chat API models in AutoAI for RAG experiments, instead of prompt template models. These models must have chat capabilities to work in AutoAI for RAG experiments.

• Auto-deploy top pattern in AutoAI for RAG experiments

  You can now enable automatic deployment of the top-performing pattern after an AutoAI for RAG experiment completes. You can turn on auto-deployment when you set up the experiment. Auto-deployment helps reduce manual steps and further automates the experiment workflow.

• Use multiple vector indexes in AutoAI for RAG experiments

  You can now select up to 20 vector indexes for your document collection in an AutoAI for RAG experiment. During experiment setup, when you add document and evaluation sources, choose Knowledge bases and then select up to 20 connections. You can then define details for each connection, such as index name and embedding models. Using multiple indexes gives you more flexibility and can improve the quality and performance of your experiments.

• Use SQL database schemas in AutoAI for RAG experiments

  You can now choose an SQL database schema as a knowledge base in an AutoAI for RAG experiment. You can use SQL connections such as Db2, PostgreSQL, and MySQL. When using SQL sources, chunking settings are disabled, and only answer correctness metrics are available for optimization. With an SQL RAG, structured data can be retrieved directly from the relational database, which can improve answer accuracy and relevance when compared with document-based sources.

Version 12.0.0 of the watsonx.ai service includes various fixes.

For details, see What's new and changed in watsonx.ai.

Version 5.3.0 of the watsonx Assistant service includes various security fixes.

Manage new roles for watsonx BI users through IBM Software Hub Access control

As a watsonx BI user, you can now use IBM Software Hub Access control to assign roles to your team that are specific to watsonx BI. You can assign the following roles to users or groups: Business Intelligence Administrator, Business Intelligence Analyst, or Business Intelligence Consumer.

For more information, see Roles and permissions for watsonx BI.

Configure flat file service to upload files to watsonx BI

You can configure flat file service to upload files and use them as a source of data in Conversations.

For more information, see Configuring flat file service.

Integrate with IBM watsonx.data Premium

You can now easily access watsonx BI from a tile in watsonx.data Premium. By combining watsonx.data Premium and watsonx BI, you can use business intelligence tools to get deeper insights into your structured and unstructured data.

For more information, see IBM watsonx.data Premium.

Version 3.3.0 of the watsonx BI service includes various security fixes.

Version 5.3.0 of the watsonx Code Assistant service includes various security fixes.

For details, see What's new and changed in watsonx Code Assistant.

Version 5.3.0 of the watsonx Code Assistant for Red Hat Ansible Lightspeed service includes various fixes.

For details, see What's new and changed in watsonx Code Assistant for Red Hat Ansible Lightspeed.

Transform COBOL files to JAVA by using SQLite

You can now transform COBOL files to JAVA by using SQLite as a local alternative to the Db2 on Cloud instance with certain limitations. For more information, see Transform COBOL files to JAVA by using SQLite.

Bulk generation provides better synchronization with Red Hat extension's Outline view

The bulk generation feature now operates in the single-threaded mode, translating one program at a time for better synchronization with the Red Hat extension's Outline view.

Translate large paragraphs of COBOL code

You can now translate large paragraphs of COBOL code because the default timeout is now set to 10 minutes across all workflows.

Version 5.3.0 of the watsonx Code Assistant for Z service includes various fixes.

For details, see What's new and changed in watsonx Code Assistant for Z.

Version 2.7.0 of the watsonx Code Assistant for Z Agentic service includes various fixes.

For details, see What's new and changed in watsonx Code Assistant for Z Agentic.

Version 2.8.0 of the watsonx Code Assistant for Z Code Explanation service includes various fixes.

For details, see What's new and changed in watsonx Code Assistant for Z Code Explanation.

Version 2.8.0 of the watsonx Code Assistant for Z Code Generation service includes various fixes.

For details, see What's new and changed in watsonx Code Assistant for Z Code Generation.

Version 2.3.0 of the watsonx.data service includes various fixes.

Version 2.3.0 of the watsonx.data Premium service includes various fixes.

Curate unstructured data with a new tool

With the new unstructured data curation tool, you can now import and analyze unstructured documents, group these documents based on the analysis results, and process the documents further based on the grouping.

You set up an analysis flow where you import metadata, detect the format and the language of documents, and classify the documents based on predefined or custom document classes. As a second step, you set up a processing flow where you transform these grouped documents, generate entities and embeddings, and create document sets and document libraries that you can then use in your gen AI projects.

The unstructured data curation tool replaces the unstructured data import and unstructured data enrichment tools.

Create SQL-based assets and data quality rules with text instead of SQL

Now you can describe the data asset or the data quality rule that you want to create in plain English and convert this text query into an SQL query. You can then run the generated query to create the asset or the rule.

Tech preview This is a technology preview and is not supported for use in production environments.

Disable certain generative AI capabilities for selected projects

Even if the product is installed with generative AI capabilities, you might not want to use these capabilities in all of your projects. You can now disable these capabilities per project. In projects where the capabilities are disabled, you can't work with natural language queries to create SQL-based assets and data quality rules. In addition, LLM-based name, description, or term generation and term assignment in metadata enrichment are disabled.

Define catalog-specific custom properties for assets

You can now restrict custom properties for assets to a specific catalog. By using catalog-specific custom properties, you can more effectively display values that pertain only to selected domains and ensure that the right information is available to the right users.

To list custom properties that are restricted by a given catalog, use the sort by scope option and scroll down to the items for the catalog that you're interested in.

Manage columns for catalogs

You can now select which columns to display in the asset listing grid by clicking Manage columns on the catalog page. Select your columns, reorder them if necessary, and save your preferences to keep the information that is most relevant for you readily available. For example, you can modify the view to show you a list of assets with the display name, owners, and date added columns only.

Optimize term assignment

With the new tuning options for term assignment, now you can influence the weighting of term suggestions for better precision or recall.

Import primary keys and foreign keys and visualize them in Relationship Explorer

Import primary keys and foreign keys with metadata import instead of metadata enrichment. After import, you can access the associated relationships through the RHS panel and Relationship Explorer.

Versioning of governance artifacts

Track historical changes for the artifacts, schedule new versions to be published in the future, and restore or archive previous versions with the new Versions panel.

Export data lineage to Collibra

You can now export data lineage and view it in Collibra. If you transfer lineage information into Collibra data governance platform, you can see a comprehensive view of your data flows and dependencies within your governance framework.

Starting parents are introduced in the data lineage graph

When you select an asset to be a starting asset in the lineage, all assets that are higher in the hierarchy are marked as starting parents. Also, all child assets of the selected asset are marked as starting assets. This distinction clarifies which assets are selected as the starting points for the lineage.

Disable data lineage for the unstructured data flows

Data lineage is generated for Unstructured Data Integration and unstructured data curation flows by default. You can disable the lineage generation for unstructured data to control when lineage is created.

Create and access data contracts in Open Data Contract Standard v3

Streamline your management of data contracts by using Open Data Contract Standard v3 (ODCS v3) format in Data Product Hub

• Producers: You can now create data contracts in ODCS v3format. Create contracts from scratch or by using a predefined template.
• Consumers: You can access and review data contracts directly in Data Product Hub or download them in YAML format, along with any associated test status information.

This optimized process enhances collaboration, ensures data quality, and enhances trust between producers and consumers.

Deliver data products from Microsoft Azure Databricks

You can now subscribe to a data product that is created in Azure Databricks by using the Access in Azure Databricks delivery method. Consumers can directly access Azure Databricks resources. After delivery of the data products, consumers see details on how to access the specific resources in Azure Databricks.

Deliver data assets to a project by using the access in watsonx.data delivery method

You can now choose to import data product assets to a project by using the access in watsonx.data delivery method.

Manage and view data product reviews

Consumers can now create, edit, and delete reviews of data products. Producers cannot manage reviews.

Version 2.3.0 of the watsonx.data intelligence service includes various fixes.

For details, see What's new and changed in watsonx.data intelligence.

Model groups in Governance Console are synchronized with watsonx.governance

Model groups in Governance Console are now synchronized with AI use cases in watsonx.governance.

When you create or update a model group in Governance Console, your changes are synchronized with the AI use case approaches in watsonx.governance. Similarly, when you create or update an approach in watsonx.governance, your changes are synchronized with the model groups in Governance Console.

The synchronization of model groups and approaches helps to ensure consistent tracking and governance of AI models.

You can turn this feature on or off in Configuration and settings.

Migrate assets to inventories in watsonx.governance

You can now migrate AI use cases and external models to inventories by using a command-line tool. Previously, these assets were stored in the platform assets catalog or other catalogs.

Evaluate and govern prompt templates that use chat as the input

You can now evaluate and govern prompt templates that use chat as the input. To get started, track the prompt template in a use case.

Note the following restrictions for prompt templates that use chat:

• You can evaluate the prompt templates only in a production deployment space.
• The evaluation of feedback data is not supported.
• The prompt templates are evaluated only on the payload data, which gets scored against the deployment endpoint. Therefore, metrics that compute values based on the similarity between reference texts and predictions, such as BLEU, ROUGE, and so on, are not evaluated.
• Only the GENAIQ monitor is supported.

Use Guardrail Manager to manage AI guardrails

Configure the predefined guardrails, select actions, and manage them in an inventory through the UI by using the new Guardrail Manager. You can also manage different guardrail configurations in Guardrail Manager. Your guardrail configurations are stored in your inventory, and you can share the inventory with others.

Version 2.3.0 of the watsonx.governance service includes various fixes.

For details, see What's new and changed in watsonx.governance.

Enter new input types and customization in forms

You can now use forms in user activities to organize multiple types of interactions within a structured layout. You can complete the following tasks:

• Add a File upload field to collect documents or images.
• Edit column labels in Multi choice and Single choice fields.
• Use the Field option to display static information during a chat.

For details, see Forms in user activities.

Use chat session context as input to improve agentic workflows

By using chat session content, you can have improved responses, reduced repetitive inputs, and context-aware workflows. Chat session context provides context from the chat as input to the prompt. You can enable the chat session context and get the last five conversations from the chat history. These conversations are converted to form fields as input to the prompt. If the conversation mapping fails, the user is prompted to provide input in the chat. For details, see Using the chat session context as input.

Handle interruptions in voice conversations by using VAD settings

You can define how your voice agent responds to user interruptions. By using Voice Activity Detection (VAD) setting, you can optimize the responsiveness and accuracy of speech recognition by identifying active voice segments and ignoring silence or background noise. VAD reduces latency and improves the user experience in voice-driven workflows. For details, see Configuring voice settings for agents.

Configure ElevenLabs TTS for natural voice output

You can configure ElevenLabs as a text-to-speech (TTS) provider in an agent builder to create voice agents with natural, human-like speech. Add your own API key, choose voice settings like speed and stability, and preview voices before you apply them. For details, see Configuring voice settings for agents.

Version 7.0.0 of the watsonx Orchestrate service includes various fixes.

For details, see What's new and changed in watsonx Orchestrate.

Architects and developers can use watsonx Code Assistant for Z Understand to work with generative AI to understand and analyze your mainframe applications. With watsonx Code Assistant for Z Understand, you can use a chat interface and access different data sources to get application insights and discover business rules.

IBM watsonx.data integration provides unified tools that you can use to transform, integrate, and observe your data. You can use a range of diverse data integration styles, such as streaming, replication, observability, and bulk or batch processing.

• Version 4.14.0 or later fixes
• Version 4.16.4 or later fixes
• Version 4.17.0 or later fixes
• Version 4.18.6 or later fixes
• Version 4.19.0 or later fixes
• Version 4.20.0 or later fixes

Starting in Version 5.3, many services in IBM Software Hub use Helm for installation and upgrade. For services that support Helm, a cluster administrator must create cluster-scoped resources, such as custom resource definitions, cluster roles, cluster role binding, and webhooks. This change gives cluster administrators more insight into the cluster-scoped resources that are required for each service.

The install-components command replaces the apply-olm and apply-cr commands. The command ensures that the required operators are created before the custom resources are created.

IBM Software Hub Version 5.3 uses the Red Hat OpenShift Container Platform cert-manager Operator.

If you are upgrading to IBM Software Hub Version 5.3, ensure that you migrate to the Red Hat OpenShift Container Platform cert-manager Operator before you upgrade IBM Software Hub.

For details, see Migrating from the IBM Certificate manager to the Red Hat OpenShift Certificate manager

The cpd-cli manage setup-instance command is deprecated. Use the new cpd-cli manage install-components command to install and upgrade the IBM Software Hub platform.

For details, see:

Installation

Installing the required components for an instance of IBM Software Hub

Upgrade from 5.1

Upgrading IBM Software Hub

Upgrade from 5.2

Upgrading IBM Software Hub

The apply-olm command is replaced by the install-components command.

The apply-cr command is replaced by the install-components command.