NGC Catalog
CLASSIC
Welcome Guest
Collections
NVIDIA AI - End-to-End AI Development & Deployment

NVIDIA AI - End-to-End AI Development & Deployment

For contents of this collection and more information, please view on a desktop device.
Logo for NVIDIA AI - End-to-End AI Development & Deployment
Description
This is a collection of performance-optimized frameworks, SDKs, and models to build Computer Vision and Speech AI applications.
Curator
NVIDIA
Modified
March 14, 2025
Containers
Sorry, your browser does not support inline SVG.
Helm Charts
Sorry, your browser does not support inline SVG.
Models
Sorry, your browser does not support inline SVG.
Resources
Sorry, your browser does not support inline SVG.

This collection provides performance-optimized frameworks, SDKs, and pre-trained models for AI practitioners to develop and deploy their solutions on any GPU-accelerated on-prem, cloud, and edge systems.

AI Tools Included in this Collection

NVIDIA releases a new version every month for many of the NVIDIA built AI software, updated with optimized libraries, delivering higher training and inference performance on the same GPU-powered system.

  • Deep Learning Frameworks: Updated monthly, PyTorch and TensorFlow containers are optimized for GPU acceleration, and contain a validated set of libraries that enable and optimize GPU performance. These containers also contain software for accelerating ETL, training, and inference.
  • RAPIDS: Accelerates end-to-end data science and analytics pipelines entirely on GPUs.
  • TensorRT: Takes a trained network and produces a highly optimized runtime engine that performs inference for that network.
  • TAO: A python-based AI toolkit for taking purpose-built pre-trained AI models and customizing them with your own data. Add all 3 TAO containers in the entities tab.
  • Triton: An open-source software to deploy trained AI models from any framework, on any GPU- or CPU-based infrastructure in the cloud, data center, or embedded devices.
  • DeepStream: This SDK delivers a complete streaming analytics toolkit for AI based video and image understanding and multi-sensor processing.
  • RIVA: A GPU-accelerated SDK for building speech applications that are customized for your use case and deliver real-time performance. Include RIVA Clients and RIVIA Speech skills

Pre-trained Models Included in this Collection

We’ve also included a few pre-trained models for computer vision and speech for you to get started. See the Models page for the full list of models across vision, speech, healthcare, and more.

Computer Vision models

  • PeopleNet: Three class object detection network to detect people in an image. https://catalog.ngc.nvidia.com/orgs/nvidia/teams/tao/models/peoplenet
  • ActionRecognitionNet: 5 class network to recognize what people are doing in an image. https://catalog.ngc.nvidia.com/orgs/nvidia/teams/tao/models/actionrecognitionnet
  • TafficCamNet: Four class object detection network to detect cars in an image. https://catalog.ngc.nvidia.com/orgs/nvidia/teams/tao/models/trafficcamnet

Speech Models

  • English Citrinet ASR: Model trained on ASR set 3.0 to transcribe segments of audio to text. https://catalog.ngc.nvidia.com/orgs/nvidia/teams/tao/models/speechtotext_en_us_citrinet
  • Conformer-CTC: Transcribes speech in lower case English alphabet along with spaces and apostrophes. https://catalog.ngc.nvidia.com/orgs/nvidia/teams/nemo/models/stt_en_conformer_ctc_medium
  • FastPitch: Speech synthesis model trained on female English speech. https://catalog.ngc.nvidia.com/orgs/nvidia/teams/nemo/models/tts_en_fastpitch

License

By pulling and using the containers and models, you accept the terms and conditions of the applicable End User License Agreement.