Create Custom AI Models With the NVIDIA TAO

NVIDIA TAO Version 5.2: What’s New

NVIDIA TAO delivers a low-code, open-source AI framework to accelerate vision AI model development suitable for all skill levels—from beginners to expert data scientists. Developers can now use the power and efficiency of transfer learning to achieve state-of-the-art accuracy and production-class throughput in record time with adaptation and optimization.

This latest release of NVIDIA TAO is leaded with groundbreaking features to boost your AI model development:

  • Export model in standard ONNX format and deploy on any platform—GPUs, CPUs, MCUs, DLAs, etc.
  • State-of-the-art vision Transformers for better accuracy and robustness against image corruption and noise
  • The ability to annotate segmentation masks with new AI-assisted data annotation capabilities
  • Automatic fine-tuning of hyperparameters with AutoML on new vision Transformers
  • New computer vision pre-trained models for optical character detection and recognition, siamese inspection, and more
  • Update of 5.1
    • Fine-Tune large vision Foundational models like CLIP and NV-DINOv2 for custom CV Tasks
    • Vision Transformer based Siamese network for visual defect classification and segmentation
  • Update of 5.2
    • Run 0-Shot inference for Panoptic Segmentation using ODISE
    • Create custom models for 3D object poses using CenterPose
    • Maximize Inference throughput for ViTs with FasterViT

Download TAO Download from Github Repo Try on LaunchPad

Production-Ready Vision AI

NVIDIA TAO is also available as a part of NVIDIA AI Enterprise, an end-to-end, secure, cloud-native AI software platform optimized to accelerate enterprises to the leading edge of AI.

With enterprise-grade security, stability, manageability, and support, NVIDIA AI Enterprise speeds time to value while mitigating the potential risks of open-source software, ensuring business continuity and a reliable platform for running mission-critical AI applications.

Benefits of using TAO with NVIDIA AI Enterprise:

  • Access to exclusive foundation models for vision AI. These models can be fine-tuned for custom Vision AI tasks using TAO
  • Validation and integration for NVIDIA AI open-source software
  • Access to AI solution workflows to speed time to production
  • Certifications to deploy AI everywhere
  • Enterprise-grade support, security, manageability, and API stability to mitigate potential risks of open source software

Learn More

Get Started With TAO on Google Colab

Access the Latest in Vision AI Model Development Workflows with NVIDIA TAO

The next version of TAO supercharges vision AI development for practically any developer, in any service and on any device

Read the Blog

Transforming Industrial Defect Detection with NVIDIA TAO and Vision AI Models

Learn how to enhance industrial defect detection accuracy using NVIDIA TAO and state-of-the-art vision transformer models.

Read the Blog

Train Character Detection and Recognition Models with NVIDIA TAO

Learn how to train and deploy a custom optical character detection and recognition model using NVIDIA TAO and NVIDIA Triton.

Read the Blog -
Part 1 |   Part 2

Developer Starter Resources

Training Notebooks & Containers

Sample Deployment Applications

To convert TAO model (etlt) to an NVIDIA TensorRT™ engine for deployment with DeepStream, select the appropriate TAO-converter for your hardware and software stack.

Additional Resources

Product Support

Ethical AI

NVIDIA platforms and application frameworks enable developers to build a wide array of AI applications. Consider potential algorithmic bias when choosing or creating the models being deployed. Also, work with the model’s developer to ensure that it meets the requirements for the relevant industry and use case; that the necessary instruction and documentation are provided to understand error rates, confidence intervals, and results; and that the model is being used under the conditions and in the manner intended.