> This page is for version 3.3.0 (default).
> For other versions, use one of these documentation indexes:
> - 3.3.0 (default): https://docs.nvidia.com/vss/3.3.0/llms.txt
> - Dev: https://docs.nvidia.com/vss/dev/llms.txt

> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://docs.nvidia.com/vss/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.nvidia.com/vss/_mcp/server.

# Release Notes

## Version 3.3.0

### Features

* Deployment profiles:

  * Added Warehouse Helm charts for 2D, 3D, and MV3DT profiles, with optional agents in the 2D chart.
  * Added support for RTX PRO 4500 Blackwell and GB300 hardware profiles.

* Launchable and Skills:

  * Added Warehouse deployment support to the Build Vision AI skill, covering 2D, 3D, MV3DT, and Auto Calibration profiles, Kafka / Redis variants, and minimal / extended configurations.
  * Added the `vss-deploy-warehouse-helm` skill and a standalone `compute_stream_cap.py` helper to set stream limits based on GPU hardware.
  * Updated Behavior Analytics skills for the Build Vision Agent to deploy standalone and combined Search + Alerts workloads, with guidance for proximity alerts, FOV counting, ROI bounding-box overlap, object speed, and people counting.
  * Updated the Warehouse Brev launchable notebook for the refactored Docker Compose configuration and unified Auto Calibration profile.

* Microservices:

  * Behavior Analytics:

    * Added `behaviorEmitOnce` to emit a final behavior when a track ends, with per-sensor event-time handling, behavior holdback, and shutdown flushing. Real-time emission remains the default, and `isComplete` identifies completed behaviors.
    * Added configurable ROI ENTRY / EXIT detection based on bounding-box overlap.

  * Auto Calibration: Added support for image-mode calibration

  * Video IO & Storage (VIOS): Added DASH live-stream playback as an alternative to WebRTC.

  * MV3DT: Integrated ReID embeddings with Kafka and Elasticsearch ingestion support.

* Models - Core AI models for perception and analytics:

  * [Sparse4D - 3D Warehouse Models v2.3 and v3.0](/system-components/models/sparse4d-3d-multi-camera) - Updated models for synthetic and real-world Warehouse datasets, respectively. `DATASET_TYPE` selects the matching model, anchor, labels, and TensorRT engine configuration.
  * Perception models are now downloaded from the NGC TAO registry during deployment instead of being bundled in the Warehouse app-data archive.

### Improvements

* Docker Compose:

  * Centralized NvStreamer, VSS Configurator, and Video Analytics API services and removed duplicated profile-specific configurations and Configurator initialization services.
  * Unified the Warehouse Auto Calibration profile to deploy NvStreamer, VIOS, Auto Calibration, and VSS Configurator together.
  * Reorganized `.env` and `overrides.env` configuration and updated Warehouse profile-selection defaults.

* Microservices:

  * Behavior Analytics:

    * Combined Search and Alerts in a single Behavior Analytics main application script.
    * Refactored restricted-area and confined-area violation detection and ROI information handling, simplified trajectory and state management, tuned default pipeline worker counts, and improved worker-lifecycle and fatal-error reporting.
    * Enabled proximity, confined-area, and restricted-area incidents in the Warehouse 3D profile.
    * Updated ROI output to report `restrictedAreaViolation=false` when no objects are present.
  * Video Analytics API: Added readiness checks for Elasticsearch connectivity, the insertion-timestamp ingest pipeline, and required Kafka topics before starting the HTTP listener and accepting calibration uploads. Redis deployments ignore Kafka configuration.
  * Logstash:

    * Replicated Logstash ingestion pipelines and updated consumer groups per Kafka topic, with worker, batch, heap, and resource tuning to reduce Elasticsearch ingestion delays.
    * Added the Redis Stream input plugin source code, build documentation, and tests to the repository.
    * Improved Redis Stream plugin connection-pool lifecycle, graceful shutdown, duplicate-message handling, acknowledgment, and protobuf descriptor loading.

* Models and hardware configuration:

  * Updated the maximum number of streams per GPU for RT-DETR and Sparse4D models.
  * Increased the RT-DETR Person-class detection limit from 20 to 50.
  * Added batch-size suffixes to TensorRT engine filenames to prevent incompatible engine reuse.
  * Tuned RTVI-VLM memory utilization and context length for RTX PRO 4500 Blackwell deployments.

### Fixed Issues

* Alerting and RTVI-VLM:

  * Fixed real-time alerts failing to initialize when Elasticsearch shards were not yet ready.
  * Fixed missing real-time VLM incidents caused by Redis message-field configuration.
  * Fixed RTVI-VLM inference failures caused by multimodal processor caching and request serialization.

* Behavior Analytics: Fixed validation of violation-duration thresholds and expiration windows to accept fractional seconds.

* SDRC: Fixed a startup race with RTVI-CV by waiting for perception readiness before adding streams.

* Video IO & Storage (VIOS):

  * Fixed incorrect 3D bounding-box overlays caused by projecting boxes behind the camera.
  * Fixed missing overlays in downloaded recordings when Elasticsearch throttles metadata requests.

## Version 3.2.1

### Features

* Models - Core AI models:

  * [CR](/vision-agent/agent-workflows/industry-specific-examples/warehouse-operations-blueprint/development-workflow/model-finetuning/cosmos-reasoner) - Updated Cosmos3 Nano Reasoner (NIM) as the default VLM for Warehouse Blueprint.

### Improvements

* VSS Configurator: Fixed an issue with live RTSP stream configuration.

## Version 3.2.0

Initial GitHub release of the VSS Warehouse Blueprint.

### Features

* New profiles introduced:

  * MV3DT profile.
  * Auto Calibration profile.

* Launchable and Skills:

  * Warehouse deploy launchable notebook.
  * Warehouse deploy and debug skills.
  * Behavior Analytics standalone deploy skill.
  * Video Analytics API standalone deploy skill.

* Microservices:

  * RTVI-VLM always-on alerts enabling Load Quality, PPE, Spillover, and Pathway / Unexpected Obstructions.

  * Replacement of local VLM with RTVI-VLM.

  * Auto Calibration:

    * Support for 3D profile deployment.
    * Support for image coordinates.
    * Support for RTSP as input.

  * Behavior Analytics: Added support for dynamic configuration changes.

  * Added dynamic configuration support in Video Analytics API to store and audit configuration changes of Behavior Analytics.

  * Spatial AI Data Utils:

    * Enable region and group assignment to global ROIs and tripwires.
    * Add evaluation tools for 3D IoU-based tracking evaluation.
    * Implement data validation and evaluation for 3D inference data.
    * Add visualization tools.

  * VSS Configurator:

    * Support toggling between Kafka and Redis configurations.
    * Support editing JSON configuration files.

  * SDR: Auto re-addition of streams on Docker restart for perception services.

### Improvements

* Docker Compose: Refactored to improve modularity of microservices.

* Microservices:

  * Video Analytics API: Retry mechanism for Kafka broker, Elasticsearch, and related services.

  * Behavior Analytics: Retry mechanism for Kafka broker, Elasticsearch, Redis, MQTT, and related services.

  * Spatial AI Data Utils: Improved camera grouping algorithms.

  * VSS Configurator:

    * Added backup and cleanup support for `calibration.json`.
    * Ensured the calibration file schema is followed after VSS Configurator operations.

  * Alerting service:

    * VLM verification supports JSON output from VLM.
    * Improved VLM verification accuracy for Near Miss Violation by tuning prompts.

  * VIOS:

    * Added top-view image visualization support on Thor and Spark platforms.
    * Improved visualization of 3D bounding boxes by using individual camera timestamps when present.

* Models - Core AI models for perception and analytics:

  * [RT-DETR - 2D Warehouse Model v1.0.2](/system-components/models/rt-detr-2d-warehouse-detection) - Model for 2D Single-Camera Object Detection.
  * [Sparse4D - 3D Warehouse Model v2.2](/system-components/models/sparse4d-3d-multi-camera) - Model for 3D Multi-Camera Object Detection and Tracking.

## Version 3.1.0

### Features

* Introduced Minimal profile to deploy the minimal set of microservices to run on a machine with smaller resource footprint

### Improvements

* Microservices

  * Perception:
    * Perception microservice uses DeepStream SDK v9.0 to perform object detection and tracking.
    * Performance improvement for IGX Thor.
  * Behavior Analytics: Enhanced support for image coordinates based analytics.
  * Agents: Improved the multi-turn conversation capabilities.

* Models - Core AI models for perception and analytics
  * [RT-DETR - 2D Warehouse Model v1.0.1](/system-components/models/rt-detr-2d-warehouse-detection) - Model for 2D Single-Camera Object Detection.
  * [Sparse4D - 3D Warehouse Model v2.1](/system-components/models/sparse4d-3d-multi-camera) - Model for 3D Multi-Camera Object Detection and Tracking.

### Fixed Issues

* Video IO & Storage (VIOS): Fixed issues with storage service.
* VSS Configurator: Fixed issues with live RTSP stream configuration.

## Version 3.0.0

Initial release of the VSS Warehouse Blueprint.

### Features

* [Warehouse Blueprint Profiles](/vision-agent/agent-workflows/industry-specific-examples/warehouse-operations-blueprint/development-workflow/deployment-customization) - Layered architecture profiles combining microservices, blueprints, and AI agents for 2D and 3D applications

  * [2D Vision AI Profile](/vision-agent/agent-workflows/industry-specific-examples/warehouse-operations-blueprint/development-workflow/deployment-customization/2d-vision-ai-profile) - 2D single-camera detection (using RTDETR model), tracking, and analytics
  * [2D Vision AI with Agents Profile](/vision-agent/agent-workflows/industry-specific-examples/warehouse-operations-blueprint/development-workflow/deployment-customization/2d-vision-ai-with-agents-profile) - 2D single-camera detection (using RTDETR model), tracking, analytics, and VSS agent integration
  * [3D Vision AI Profile](/vision-agent/agent-workflows/industry-specific-examples/warehouse-operations-blueprint/development-workflow/deployment-customization/3d-vision-ai-profile) - 3D multi-camera detection (using Sparse4D model), tracking, and analytics

* Microservices - Core components including video I/O and storage, behavior analytics, RESTful APIs, operations UI, AI agents and Auto Calibration.

  * [Video IO & Storage (VIOS)](/system-components/middleware/video-io-storage-vios) – Microservice for ingesting camera streams (RTSP, WebRTC), managing video recording, playback, and persistent storage
  * [Behavior Analytics](/system-components/analytics-microservices/behavior-analytics) – Real-time detection and analysis of warehouse events including object tracking, zone monitoring, and behavior pattern recognition
  * [Video Analytics APIs](/system-components/middleware/api-gateway-mcp/video-analytics-api) – RESTful APIs providing programmatic access to analytics data, event streams, and video workflow management
  * **Reference Video Analytics UI** – Web-based dashboard for real-time monitoring, incident review, camera management, and analytics visualization
  * [VSS Agent](/vision-agent/agent-workflows/industry-specific-examples/warehouse-operations-blueprint/microservice-configuration/agents) – Natural language interface powered by Nemotron Nano 9B v2 enabling conversational queries about incidents and automated report generation

* Models - Core AI models for perception and analytics

  * [RT-DETR - 2D Warehouse Model v1.0](/system-components/models/rt-detr-2d-warehouse-detection) - Model for 2D Single-Camera Object Detection.
  * [Sparse4D - 3D Warehouse Model v2.0](/system-components/models/sparse4d-3d-multi-camera) - Model for 3D Multi-Camera Object Detection and Tracking.
  * [CR](/vision-agent/agent-workflows/industry-specific-examples/warehouse-operations-blueprint/development-workflow/model-finetuning/cosmos-reasoner) - Reasoning Vision Language Model (VLM) enabling multi-step reasoning across visual and textual inputs.

## Known Limitations

See [Known Limitations](/vision-agent/agent-workflows/industry-specific-examples/warehouse-operations-blueprint/operations-monitoring-troubleshooting-faq-known-limitations/known-limitations) for current constraints and workarounds.

## Related Topics

* [Troubleshooting Guide](/vision-agent/agent-workflows/industry-specific-examples/warehouse-operations-blueprint/operations-monitoring-troubleshooting-faq-known-limitations/troubleshooting-guide)
* [Blueprint Profiles](/vision-agent/agent-workflows/industry-specific-examples/warehouse-operations-blueprint/development-workflow/deployment-customization)