> This page is for Self-Hosted, version v3.0 (default).
> For other versions, use one of these documentation indexes:
> - v3.0 (default): https://docs.nvidia.com/run-ai/self-hosted/v3.0/llms.txt
> - v2.26: https://docs.nvidia.com/run-ai/self-hosted/v2.26/llms.txt
> - v2.25: https://docs.nvidia.com/run-ai/self-hosted/v2.25/llms.txt
> - v2.24: https://docs.nvidia.com/run-ai/self-hosted/v2.24/llms.txt

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

# Videos

Explore NVIDIA Run:ai product videos, demos, and walkthroughs covering key platform workflows and capabilities.

<Cards>
  <Card title="Platform Overview" src="https://img.youtube.com/vi/V3oPZ6SsQDQ/hqdefault.jpg" imagePosition="top" href="/run-ai/self-hosted/documentation/resources/videos/platform-overview">
    Get an overview of the NVIDIA Run:ai platform.
  </Card>
  <Card title="How to Run a Jupyter Notebook Using Workspaces" src="https://img.youtube.com/vi/XwSgU-Atb_Y/hqdefault.jpg" imagePosition="top" href="/run-ai/self-hosted/documentation/resources/videos/how-to-run-a-jupyter-notebook-using-workspaces">
    Launch an interactive Jupyter Notebook environment using workspaces.
  </Card>
  <Card title="How to Run a Standard Training Job" src="https://img.youtube.com/vi/A1mkRye5_To/hqdefault.jpg" imagePosition="top" href="/run-ai/self-hosted/documentation/resources/videos/how-to-run-a-standard-training-job">
    Submit and monitor a standard training workload.
  </Card>
  <Card title="How to Run a Nemo Training Job" src="https://img.youtube.com/vi/WC4Dv-U6NX4/hqdefault.jpg" imagePosition="top" href="/run-ai/self-hosted/documentation/resources/videos/how-to-run-a-nemo-training-job">
    Run a training job using NVIDIA NeMo.
  </Card>
  <Card title="How to Deploy NVIDIA NIM Inference" src="https://img.youtube.com/vi/pRbGAS6-X54/hqdefault.jpg" imagePosition="top" href="/run-ai/self-hosted/documentation/resources/videos/how-to-deploy-nvidia-nim-inference">
    Deploy an inference workload using NVIDIA NIM.
  </Card>
  <Card title="How to Deploy Dynamo Inference Pipelines" src="https://img.youtube.com/vi/6u6Ku3Cx7aI/hqdefault.jpg" imagePosition="top" href="/run-ai/self-hosted/documentation/resources/videos/how-to-deploy-dynamo-inference-pipelines">
    Deploy NVIDIA Dynamo inference pipelines on Run:ai.
  </Card>
  <Card title="How to Allocate GPU Resources Across Teams" src="https://img.youtube.com/vi/UFArO8_aAYw/hqdefault.jpg" imagePosition="top" href="/run-ai/self-hosted/documentation/resources/videos/how-to-allocate-gpu-resources-across-teams">
    Configure quota and resource allocation across departments and projects.
  </Card>
  <Card title="How to Run a Custom Inference Workload" src="https://img.youtube.com/vi/kCb0n9A3sbc/hqdefault.jpg" imagePosition="top" href="/run-ai/self-hosted/documentation/resources/videos/how-to-run-a-custom-inference-workload">
    Deploy a custom inference server on NVIDIA Run:ai.
  </Card>
  <Card title="Topology-Aware GPU Scheduling" src="https://img.youtube.com/vi/8Ws2DBWo40A/hqdefault.jpg" imagePosition="top" href="/run-ai/self-hosted/documentation/resources/videos/topology-aware-gpu-scheduling">
    Use network topology-aware scheduling to accelerate distributed workloads.
  </Card>
  <Card title="Deploy and Manage NVIDIA RAG Blueprint" src="https://img.youtube.com/vi/VzlTZue-ISc/hqdefault.jpg" imagePosition="top" href="/run-ai/self-hosted/documentation/resources/videos/deploy-and-manage-nvidia-rag-blueprint">
    Deploy and manage the NVIDIA RAG Blueprint on Run:ai.
  </Card>
  <Card title="How to Create an AI Application" src="https://img.youtube.com/vi/1o72Z4vDj0o/hqdefault.jpg" imagePosition="top" href="/run-ai/self-hosted/documentation/resources/videos/how-to-create-an-ai-application">
    Create and publish an AI application on NVIDIA Run:ai.
  </Card>
  <Card title="Over-Quota GPU Usage with Fairness and Preemption" src="https://img.youtube.com/vi/i7ufKl6_8jc/hqdefault.jpg" imagePosition="top" href="/run-ai/self-hosted/documentation/resources/videos/over-quota-gpu-usage-with-fairness-and-preemption">
    Understand how over-quota, fairness, and preemption work together.
  </Card>
  <Card title="How to Run a Distributed Training Job" src="https://img.youtube.com/vi/QAPfsGE5iO0/hqdefault.jpg" imagePosition="top" href="/run-ai/self-hosted/documentation/resources/videos/how-to-run-a-distributed-training-job">
    Submit a distributed training workload across multiple nodes.
  </Card>
  <Card title="NVIDIA Run:ai on NVIDIA NVL72" src="https://img.youtube.com/vi/DVdd0OapgDU/hqdefault.jpg" imagePosition="top" href="/run-ai/self-hosted/documentation/resources/videos/nvidia-run-ai-on-nvidia-nvl-72">
    Run AI workloads on GB200 NVL72 hardware with NVIDIA Run:ai.
  </Card>
  <Card title="Simplify AI Workload Submission with Workload Assets" src="https://img.youtube.com/vi/blFfTRjw-xo/hqdefault.jpg" imagePosition="top" href="/run-ai/self-hosted/documentation/resources/videos/simplify-ai-workload-submission-with-workload-assets">
    Use workload assets to streamline and standardize workload submission.
  </Card>
  <Card title="Create and Reuse Workload Templates" src="https://img.youtube.com/vi/OZ_Wg5-d2nY/hqdefault.jpg" imagePosition="top" href="/run-ai/self-hosted/documentation/resources/videos/create-and-reuse-workload-templates">
    Build reusable templates to speed up workload submission.
  </Card>
  <Card title="Monitor GPU Usage and Platform Health" src="https://img.youtube.com/vi/eO5L0Qm34AM/hqdefault.jpg" imagePosition="top" href="/run-ai/self-hosted/documentation/resources/videos/monitor-gpu-usage-and-platform-health">
    Monitor GPU utilization and platform health across your cluster.
  </Card>
</Cards>