Cloudera Docs

What's New

Major features and updates for the Cloudera AI data service.

April 21, 2026

Release notes and fixed issues for Cloudera AI Workbench 2.0.55-h1000-b5, Cloudera AI Registry 1.13.0-h1000-b5, and Cloudera AI Inference service 1.10.0-h1000-b9 provide functional updates across these components and resolutions for identified bugs.

New Features and Improvements

Cloudera AI Inference service
  • You can now generate credentials to authenticate application requests or test model endpoints directly from the Model Endpoint Details page. For more information, see Authenticating with model endpoints.
  • Model endpoint creation and editing UI has been redesigned into a guided, wizard-based flow that unifies model version selection, resource profiles, autoscaling, and review in a single experience. For more information, see Creating a Model Endpoint using UI.
  • When deploying Hugging Face models, you can now use an optional task selector in the UI to explicitly define operations such as text generation, embedding, or reranking. This selection is persisted in the deployment payload. For more information, see Creating a Model Endpoint using UI.
  • Cloudera AI Inference service now supports the deployment of the NVIDIA Magpie TTS Multilingual NIM runtime for high-quality text-to-speech (TTS) workloads. Using the new TEXT_TO_SPEECH task, you can generate multilingual synthetic speech in both offline (WAV) and streaming (LPCM) formats.
  • Cloudera AI Inference service now supports the NVIDIA Whisper Large v3 NIM for high-accuracy automatic speech recognition (ASR) workloads. By utilizing the new riva_asr_whisper_large_v3_v140 runtime, users can deploy the latest v1.4.0 Whisper image to significantly improve transcription performance. For backward compatibility, existing deployments using Whisper versions 1.3.0 and 1.3.1 remain fully supported and unchanged.
  • Cloudera AI Inference service now supports NVIDIA Parakeet 1.1B CTC EN-US NIM v1.4.0 for high-performance English speech-to-text workloads. Compared with larger models such as Whisper large-v3, Parakeet 1.1B is optimised for English-only use cases with a smaller model footprint, faster inference, and strong transcription accuracy for EN-US deployments. Users can deploy this model as a Riva ASR NIM through the Cloudera AI Inference service model catalog.
Cloudera AI Registry
  • Cloudera has introduced a public Model Hub catalog, a curated external repository that allows you to browse and evaluate recommended NVIDIA NIMs and Hugging Face models before importing them. This standalone catalog accelerates model selection and simplifies collaboration by providing stakeholders outside of the Cloudera environment with direct visibility into supported AI assets.
Cloudera AI Workbench
  • Cloudera AI now provides an APIv2 endpoint that allows Administrators to upload and register ML Runtime add-ons. The API accepts the JSON ML Runtime add-on repository file, validates its contents, and then initiates asynchronous loading of the add-ons. Administrators can upload add-ons through the Swagger UI or by using command-line tools such as cURL. For more information, see Uploading ML Runtime add-on repository files.

Fixed Issues

Cloudera AI Workbench
  • CVE fixes - This release includes numerous security fixes for critical and high Common Vulnerability and Exposures (CVE).
  • Fixed an issue where deploying models from the Cloudera AI Registry to Cloudera AI Workbench would fail during the build step due to an RBAC permission error. (DSE-54619)
  • Fixed an issue where resetting a team's SSH key from the Site Administration page failed to rotate the key due to a JavaScript error. (DSE-54840)
  • Fixed the GET /api/v2/workloads/executions API to ensure workload_crn and workload_execution_crn are consistently populated for all jobs, including those that are failed, skipped, or deleted. This ensures reliable telemetry export to Cloudera Observability regardless of pod creation status. (DSE-52158)
  • Fixed an issue where the Users quota table in the Workbench UI incorrectly displayed zero values for custom quotas. (DSE-50999)
  • Fixed an issue where changes to job retry settings (script failure, system failure, timed out runs, and skipped runs) were saved but not displayed correctly in the Job Settings page. You can now view and modify the saved retry conditions as expected. (DSE-51041)
  • Fixed an issue where a warning exception appeared in the browser console immediately after landing on the home page. This warning did not impact the ability to use the Quick Find feature and has now been fixed. (DSE-49110)
  • Fixed a security issue in Cloudera AIWorkbench by restricting the ports exposed by the Istio-based load balancer. (DSE-54413)
Cloudera AI Inference service
  • Fixed a security issue in Cloudera AI Inference service where the Istio ingress for model serving exposed unnecessary ports. The ingress is now strictly restricted to port 443 for secure HTTPS traffic. (DSE-54459)
Cloudera AI Registry
  • Fixed a security issue in Cloudera AI Registry by restricting the ports exposed by the Istio mesh to only those required for core functionality. (DSE-54431)

March 18, 2026

Release notes and fixed issues for Cloudera AI Workbench 2.0.55-b196, Cloudera AI Registry 1.13.0-b41, and Cloudera AI Inference service 1.10.0-b91 provide functional updates across these components and resolutions for identified bugs..

New Features and improvements

Cloudera AI Workbench
  • A new Skip Validation option has been added to skip the pre-flight validation checks before upgrading the workbench. Select this option only if validation checks are failing incorrectly.

Fixed Issue

Cloudera AI Control Plane
  • Fixed a container runtime regression (runc v1.4.0) that caused health probe failures on Azure. The Buildkit DaemonSet probe mechanism was updated to ensure stable image builds and to guarantee that the MLX secret is preserved in the monitoring namespace during workspace upgrades and restorations. (DSE-52078)
Cloudera AI Workbench
  • Fixed an issue where Workload Authorization Groups were inadvertently deleted during the restore-based upgrade workflow. (DSE-53452)

February 19, 2026

Release notes and fixed issues for Cloudera AI Workbench 2.0.55-b193, Cloudera AI Registry 1.13.0-b41, and Cloudera AI Inference service 1.10.0-b91 provide functional updates across these components and resolutions for identified bugs.

New Features / Improvements

Cloudera AI Workbench
  • Improved application responsiveness by optimizing database query performance across projects and jobs, significantly reducing page load times for environments with large datasets.
Cloudera AI Control Plane
  • Improved consistency across the administrative experience for AI Workbench, AI Registry, and the AI Inference services.
  • Added support for Amazon EKS 1.33.
  • Added support for Azure AKS 1.33.
  • Improved the reliability of Cloudera AI Inference service upgrade processes.
  • Cloudera AI now automatically performs side-by-side upgrades for incompatible environments, ensuring a seamless one-click experience with built-in rollback support in case of failure. To upgrade a workbench, Cloudera users must hold both the MLAdmin and EnvironmentAdmin roles, and those performing an upgrade must also possess the necessary cloud provider permissions to execute backup and restore operations for the underlying storage and metadata databases. For more information, see Upgrading Cloudera AI Workbenches.
Cloudera AI Inference service
  • Cloudera AI Inference service now provides a production-grade serving environment for hosting applications. Applications deployed on Cloudera AI Inference service can scale alongside Model Endpoints, providing a scalable solution for various components. For more information, see Serving Applications on Cloudera AI Inference service (Technical Preview) .
  • Cloudera AI Inference service now supports AWS on-demand capacity reservations and capacity blocks to ensure compute availability for inference workloads. For more information, see Configuring AWS on-demand capacity reservations and capacity blocks.
  • Cloudera AI now supports the deployment of Hugging Face reranking models using the API.
  • Cloudera AI now supports deploying Hugging Face embedding models using the API.
  • You can now manually specify model tasks (such as EMBED, RANK, or CLASSIFICATION) using API during deployment, enabling broader vLLM support for architectures like bertmodel or modernbertfortokenclassification that serve tasks like embedding and reranking respectively.
  • You can now manage Cloudera AI Inference service logging globally using the Serving API ConfigMap, allowing administrators to enable logging and define a storage bucket across all endpoints simultaneously for consistent data collection.
  • Added Fine Grained Authorization support. For more information, see Configuring Fine-grained Access Control.
Cloudera AI Registry
  • Cloudera AI Inference service now supports direct deployment for XGBoost, PyTorch, and TensorFlow models using the AI Registry. For more information, see Deploying Additional Model Frameworks.
  • Cloudera AI Registry now displays structured metadata and comprehensive lineage tracking (provider, model ID, and SHA) for all models imported from Hugging Face and NVIDIA NGC.

Fixed Issues

Cloudera AI Workbench
  • Resolved an issue that prevented Team Administrators from updating the team description in the UI. (DSE-48770)
  • Fixed the workload status synchronisation issue between dashboards and dashboard_pods, ensuring accurate and consistent resource usage reporting. (DSE-46977)
Cloudera AI Inference service
  • Resolved an issue where Hugging Face model IDs in inference URLs were un-sanitized, causing routing failures; inference endpoints now correctly use underscores (_) instead of slashes to align with Triton directory structures and authorizer patterns. (DSE-50543)
  • Fixed an issue where Hugging Face embedding models failed to deploy to vLLM due to an unsupported embed task key. (DSE-49993)