> This page is for version 9.2 (default).
> For other versions, use one of these documentation indexes:
> - 9.2 (default): https://docs.nvidia.com/deepstream/9.2/llms.txt
> - Archives: https://docs.nvidia.com/deepstream/archives/llms.txt

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

# DeepStream Reference Application on GitHub

## Use case applications

* [DeepStream License Plate detection and recognition](https://github.com/NVIDIA/DeepStream/tree/main/src/apps/tao_apps/apps/tao_others/deepstream_lpr_app)
* [DeepStream parallel multiple models](https://github.com/NVIDIA/DeepStream/tree/main/src/apps/reference_apps/deepstream_parallel_inference_app)
* [DeepStream text recognition with OCD/OCR models](https://github.com/NVIDIA/DeepStream/tree/main/src/apps/tao_apps/apps/tao_others/deepstream-nvocdr-app)
* [DeepStream object embedding vector generation](https://github.com/NVIDIA/DeepStream/tree/main/src/apps/tao_apps/apps/tao_others/deepstream-mdx-perception-app)
* [DeepStream pose classification](https://github.com/NVIDIA/DeepStream/tree/main/src/apps/tao_apps/apps/tao_others/deepstream-pose-classification)
* [DeepStream custom preprocessing with nvdspreprocess](https://github.com/NVIDIA/DeepStream/tree/main/src/apps/tao_apps/apps/tao_others/deepstream_custom_preprocessing_app)
* [DeepStream super resolution pipeline with buffer transferring through IPC](https://github.com/NVIDIA/DeepStream/tree/main/src/apps/reference_apps/deepstream-ipc-test-sr)
* [vLLM based VLM inference DeepStream plugin](https://github.com/NVIDIA/DeepStream/tree/main/src/apps/reference_apps/deepstream-vllm-plugin) (see [DeepStream Reference Application - deepstream-vllm-plugin](/text/DS_ref_app_vllm_plugin))

## AI models with DeepStream

* [TAO model deployment in DeepStream](https://github.com/NVIDIA/DeepStream/tree/main/src/apps/tao_apps)

## DeepStream features sample

* [Runtime source addition/removal with DeepStream](https://github.com/NVIDIA/DeepStream/tree/main/src/apps/reference_apps/runtime_source_add_delete)
* [Anomaly detection using NV Optical Flow using DeepStream](https://github.com/NVIDIA/DeepStream/tree/main/src/apps/reference_apps/anomaly)
* [MaskTracker in DeepStream](https://github.com/NVIDIA/DeepStream/tree/main/src/apps/reference_apps/deepstream-masktracker)
* [3D Body Pose (KAMA) with DeepStream](https://github.com/NVIDIA/DeepStream/tree/main/src/apps/reference_apps/deepstream-bodypose-3d)
* [Single View 3D Tracking with DeepStream](https://github.com/NVIDIA/DeepStream/tree/main/src/apps/reference_apps/deepstream-tracker-3d)
* [Multi-view 3D Tracking with DeepStream](https://github.com/NVIDIA/DeepStream/tree/main/src/apps/reference_apps/deepstream-tracker-3d-multi-view)
* [PyServiceMaker SONYC audio classifier Flow API sample](https://github.com/NVIDIA/DeepStream/tree/main/src/apps/reference_apps/pyservicemaker_sample_apps/flow_api/deepstream_audio_test_app)
* [PyServiceMaker SONYC audio classifier Pipeline API sample](https://github.com/NVIDIA/DeepStream/tree/main/src/apps/reference_apps/pyservicemaker_sample_apps/pipeline_api/deepstream_audio_test_app)
* [Save image metadata from DeepStream pipeline (Python)](https://github.com/NVIDIA-AI-IOT/deepstream_python_apps/tree/master/apps/deepstream-imagedata-multistream)
* [Generate RTSP output from the RTSP source with inferencing result (Python)](https://github.com/NVIDIA-AI-IOT/deepstream_python_apps/tree/master/apps/deepstream-rtsp-in-rtsp-out)
* [Multiple input streams and multiple separated output streams inferencing (Python)](https://github.com/NVIDIA-AI-IOT/deepstream_python_apps/tree/master/apps/deepstream-demux-multi-in-multi-out)
* [Arbitrary tiling positions and sizes configuration with nvmultistreamtiler](https://github.com/NVIDIA/DeepStream/tree/main/src/apps/reference_apps/deepstream-custom-tile-config)