NVIDIA Holoscan for Media

NVIDIA Holoscan for Media is an open reference architecture and developer toolkit for building AI-powered live media solutions on NVIDIA infrastructure.

Get StartedDocumentation


How Holoscan for Media Works

NVIDIA Holoscan for Media is an open reference architecture and developer toolkit for building AI-powered media functions and applications for software-defined live production. Developers can use reusable components, tools, and implementation patterns for GPU acceleration, high-performance networking, media transport, orchestration, control, AI processing, and application integration. The architecture provides a framework for combining these technologies on supported NVIDIA infrastructure. Teams can use the components they need to build individual applications, shared runtime environments, or distributed media functions, reducing integration complexity and moving from concept to production faster while preserving flexibility in how production systems are designed, assembled, and deployed.


Use Cases

Build a Runtime Environment for Live Media Applications

Developers at GSI or broadcast studios can leverage a Kubernetes-based reference architecture optimized for network and GPU performance. This helps them to provision and manage accelerated clusters for deploying and operating multiple live media applications.

Build a Media Processing Application

Developers can use technologies such as DeepStream, Rivermax, and reference code to build applications that handle live video ingest, GPU-accelerated processing, and AI inference. These components can support applications ranging from video transformation and analysis to AI-enabled live production workflows.

Use MXL in Media Processing Applications

Developers can use MXL, NMOS, and GPU acceleration to create software-defined media functions that exchange live media across distributed environments.


Benefits

Faster Development and Deployment

Teams can start with reusable components, tools, and reference architectures, reducing the integration work required to move from concept to production.

High-Performance Media and AI

GPU acceleration, high-performance networking, and media transport technologies support demanding live processing and AI inference workloads.

Modular, Interoperable Architecture

Developers can select the components they need, connect third-party applications and services, and reuse the same architecture across multiple media applications.


Get Started With Holoscan for Media

Developer Guide

This developer guide and the associated NGC collection provide Holoscan for Media install files and the information needed to use them.

Start Here

Local Developer Setup

This resource contains files for setting up local development environments using Kubernetes.

See Guide

Release Notes

See what’s new, improved, and available in the latest release.

See the Latest Release

More Resources

Join the NVIDIA Developer Program

Join the NVIDIA Developer Program

Sign up for the NVIDIA Developer Newsletter

Sign up for the Developer Newsletter

Join the Developer Community

Explore Developer Forums


Ethical AI

NVIDIA believes Trustworthy AI is a shared responsibility and we have established policies and practices to enable development for a wide array of AI applications. When downloaded or used in accordance with our terms of service, developers should work with their supporting model team to ensure this model meets requirements for the relevant industry and use case and addresses unforeseen product misuse. 

Please report security vulnerabilities or NVIDIA AI Concerns here.

Get started with Holoscan for Media today

Get Started