# NVIDIA Run:ai SaaS Product Documentation

Find all the product information, step-by-step guides, and reference documentation for deploying and operating NVIDIA Run:ai in SaaS cloud environments.

## New to NVIDIA Run:ai

Begin with the quick start guides to set up the platform and run your first workloads.

<table data-view="cards"><thead><tr><th></th><th data-hidden data-card-target data-type="content-ref"></th></tr></thead><tbody><tr><td><h4><i class="fa-gears">:gears:</i></h4><p><strong>Infrastructure Administrators</strong><br>Install the platform, complete the onboarding wizard, and set up cluster infrastructure</p></td><td><a href="getting-started/quick-starts/infra-admin-quick-start">infra-admin-quick-start</a></td></tr><tr><td><h4><i class="fa-sitemap">:sitemap:</i></h4><p><strong>Platform Administrators</strong><br>Configure org structure, access control, resource quotas, and governance policies</p></td><td><a href="getting-started/quick-starts/platform-admin-quick-start">platform-admin-quick-start</a></td></tr><tr><td><h4><i class="fa-laptop-code">:laptop-code:</i></h4><p><strong>AI Practitioners</strong><br>Launch workspaces, run training, and deploy inference</p></td><td><a href="getting-started/quick-starts/ai-practitioner-quick-start">ai-practitioner-quick-start</a></td></tr></tbody></table>

***

## Already Managing a Deployment

Review the latest changes and upgrade guidance before making updates to your environment.

<table data-view="cards"><thead><tr><th></th><th data-hidden data-card-target data-type="content-ref"></th></tr></thead><tbody><tr><td><h4><i class="fa-bullhorn">:bullhorn:</i></h4><p><strong>What's New</strong><br>New features, enhancements, and hotfixes</p></td><td><a href="getting-started/whats-new-for-nvidia-run-ai-saas">whats-new-for-nvidia-run-ai-saas</a></td></tr><tr><td><h4><i class="fa-circle-up">:circle-up:</i></h4><p><strong>Upgrade</strong><br>Step-by-step instructions to upgrade your cluster</p></td><td><a href="getting-started/installation/install-using-helm/upgrade">upgrade</a></td></tr><tr><td><h4><i class="fa-calendar">:calendar:</i></h4><p><strong>Product Version Life Cycle</strong><br>Plan upgrades based on support timelines</p></td><td><a href="support-policy/product-version-life-cycle">product-version-life-cycle</a></td></tr></tbody></table>

***

## Workloads (AI Lifecycle )

Step-by-step guides for running workloads at each stage of the AI lifecycle - from interactive experimentation to model training and production deployment.

<table data-view="cards"><thead><tr><th></th><th data-hidden data-card-target data-type="content-ref"></th></tr></thead><tbody><tr><td><h4><i class="fa-laptop">:laptop:</i></h4><p><strong>Workspaces (Experiment)</strong><br>Create an interactive environment for model development and experimentation</p></td><td><a href="workloads-in-nvidia-run-ai/using-workspaces/running-workspace">running-workspace</a></td></tr><tr><td><h4><i class="fa-gears">:gears:</i></h4><p><strong>Training (Standard)</strong><br>Submit a standard training workload to build and train your model</p></td><td><a href="workloads-in-nvidia-run-ai/using-training/train-models">train-models</a></td></tr><tr><td><h4><i class="fa-network-wired">:network-wired:</i></h4><p><strong>Training (Distributed)</strong><br>Run a distributed training workload across multiple nodes</p></td><td><a href="workloads-in-nvidia-run-ai/using-training/distributed-training-models">distributed-training-models</a></td></tr><tr><td><h4><i class="fa-rocket">:rocket:</i></h4><p><strong>Inference (deploy)</strong><br>Deploy and serve AI models with autoscaling and multi-node support</p></td><td><a href="workloads-in-nvidia-run-ai/using-inference/nvidia-run-ai-inference-overview">nvidia-run-ai-inference-overview</a></td></tr><tr><td><h4><i class="fa-file-code">:file-code:</i></h4><p><strong>Supported workload types (via YAML )</strong><br>Submit supported workload types (such as Dynamo) using a YAML manifest via the UI</p></td><td><a href="workloads-in-nvidia-run-ai/submit-via-yaml">submit-via-yaml</a></td></tr><tr><td><h4><i class="fa-book-open">:book-open:</i></h4><p><strong>Tutorials (end-to-end)</strong><br>Complete workflows that connect workspace → training → deployment</p></td><td><a href="tutorials/inference-tutorials">inference-tutorials</a></td></tr></tbody></table>

***

## Browse by Topic

Navigate directly to the documentation area most relevant to your role or task.

<table data-view="cards"><thead><tr><th></th><th></th><th></th><th></th><th></th><th></th><th></th></tr></thead><tbody><tr><td><h4><i class="fa-gears">:gears:</i></h4><p><strong>Install, set up and monitor</strong></p></td><td><a href="getting-started/installation/install-using-helm/system-requirements">Install SaaS</a></td><td><a href="infrastructure-setup/authentication/overview">Set authenticated access</a></td><td><a href="infrastructure-setup/advanced-setup/node-roles">Set node roles and advanced cluster configurations</a></td><td><a href="infrastructure-setup/procedures/clusters">Monitor, manage and restore clusters</a></td><td><a href="infrastructure-setup/procedures/monitoring">Monitor your platform</a></td><td></td></tr><tr><td><h4><i class="fa-sitemap">:sitemap:</i></h4><p><strong>Manage organizations and resources</strong></p></td><td><a href="../platform-management/aiinitiatives/adapting-ai-initiatives#mapping-your-organization">Map and set up your organizations</a></td><td><a href="../platform-management/aiinitiatives/adapting-ai-initiatives#mapping-your-resources">Set up and assign your resources</a></td><td><a href="../platform-management/aiinitiatives/adapting-ai-initiatives#assigning-users-to-projects-and-departments">Manage permissions</a></td><td><a href="platform-management/policies/policies-and-rules">Create and manage policies</a></td><td><a href="platform-management/monitor-performance/before-you-start">Monitor performance and health</a></td><td></td></tr><tr><td><h4><i class="fa-laptop-code">:laptop-code:</i></h4><p><strong>Build, train and deploy models</strong></p></td><td><a href="workloads-in-nvidia-run-ai/introduction-to-workloads">Learn more about workloads and workload types</a></td><td><a href="workloads-in-nvidia-run-ai/assets/overview">Prepare workload assets</a></td><td><a href="workloads-in-nvidia-run-ai/using-workspaces/running-workspace">Build your model using workspaces</a></td><td><a href="workloads-in-nvidia-run-ai/using-training/train-models">Train your model using standard or distributed training workloads</a></td><td><a href="workloads-in-nvidia-run-ai/using-inference/custom-inference">Deploy your model with inference workloads</a></td><td></td></tr><tr><td><h4><i class="fa-calendar-circle-plus">:calendar-circle-plus:</i></h4><p><strong>Scheduling and resource optimization</strong></p></td><td><a href="platform-management/runai-scheduler/scheduling/concepts-and-principles">Learn the NVIDIA Run:ai Scheduler concepts and principles</a></td><td><a href="platform-management/runai-scheduler/scheduling/how-the-scheduler-works">Understand more about how the Scheduler works</a></td><td><a href="platform-management/runai-scheduler/resource-optimization/fractions">Explore different resource optimizations</a></td><td></td><td></td><td></td></tr><tr><td><h4><i class="fa-square-terminal">:square-terminal:</i></h4><p><strong>Use the CLI</strong></p></td><td><a href="reference/cli/install-cli">Install and configure the CLI</a></td><td><a href="reference/cli/runai">See the full list of commands and examples</a></td><td></td><td></td><td></td><td></td></tr><tr><td><h4><i class="fa-rabbit-running">:rabbit-running:</i></h4><p><strong>Quick starts</strong></p></td><td><a href="workloads-in-nvidia-run-ai/using-workspaces/quick-starts/jupyter-quickstart">Run Jupyter Notebook using workspaces</a></td><td><a href="workloads-in-nvidia-run-ai/using-training/quick-starts/distributed-training-quickstart">Run your first distributed training workload</a></td><td><a href="platform-management/runai-scheduler/resource-optimization/quick-starts/dynamic-gpu-fractions-quickstart">Launch workloads with dynamic GPU fractions</a></td><td></td><td></td><td></td></tr></tbody></table>