> This page is for Multi-Tenant, version v3.0 (default).
> For other versions, use one of these documentation indexes:
> - v3.0 (default): https://docs.nvidia.com/run-ai/multi-tenant/v3.0/llms.txt
> - v2.26: https://docs.nvidia.com/run-ai/multi-tenant/v2.26/llms.txt
> - v2.25: https://docs.nvidia.com/run-ai/multi-tenant/v2.25/llms.txt
> - v2.24: https://docs.nvidia.com/run-ai/multi-tenant/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.
# Support Matrix
The support matrix outlines the verified compatibility standards for NVIDIA Run:ai multi-tenant v3.0, and applies whether you install the control plane using [Guided Installation](/run-ai/multi-tenant/documentation/installation-and-configuration/install/guided-installation) or [Manual Helm](/run-ai/multi-tenant/documentation/installation-and-configuration/install/manual-helm/preparations). To ensure a stable and performant deployment, all infrastructure components, including Kubernetes/OpenShift distributions, NVIDIA Operators, and specialized frameworks, must align with the versions specified below. Use this matrix as a validation checklist prior to performing new installations or upgrades. Once you've confirmed your versions, continue to [System Requirements](/run-ai/multi-tenant/documentation/installation-and-configuration/system-requirements) to prepare your control plane and tenant cluster infrastructure.
## Operator and Framework Versions
### Prerequisites
The following apply to each tenant cluster.
| Component | Supported Versions |
| ------------------------------------- | -------------------- |
| NVIDIA GPU Operator1 | 25.10 - 26.7.1 |
| NVIDIA DRA driver2 | 25.8 - 25.12 |
| NVIDIA Network Operator | 25.10 - 26.7 |
| Prometheus / Kube‑Prometheus Stack | 3.13.1 / 88.6.5 and above |
1 GPU Operator 26.7.1 is required to use GPU fractions, alongside CUDA driver r615 or later. GPU Operator 26.7.1 does not default to CUDA driver r615, so the driver version must be set explicitly. See [GPU Memory Fractions](/run-ai/self-hosted/documentation/platform-management/runai-scheduler/resource-optimization/gpu-fractions/memory-fractions#prerequisites).
2 Applies to a separately installed DRA driver, used when the GPU Operator runs in `ClusterPolicy` mode (the default). With GPU Operator 26.7 in `GPUCluster` mode, the GPU Operator deploys the DRA driver itself. See [NVIDIA Dynamic Resource Allocation (DRA) Driver](/run-ai/multi-tenant/documentation/installation-and-configuration/install/guided-installation/onboard-a-tenant/prerequisites-and-features#nvidia-dynamic-resource-allocation-dra-driver).
### Features
The following apply to each tenant cluster.
| Component | Supported Versions |
| ------------------------------------- | -------------------- |
| Knative Serving3 | 1.21 - 1.23 |
| Kubeflow Training Operator | 1.9.4 |
| MPI Operator | 0.6.0 - 0.8.2 |
| Leader-Worker Set (LWS) | 0.7.0 - 0.10.0 |
| NIM Operator | 3.1.x |
| Dynamo Operator | 1.5.x |
3 On OpenShift, Knative is installed through the OpenShift Serverless operator, which supports Knative 1.17 only.
## Third-Party Service Versions
Each NVIDIA Run:ai release is tested and validated against a single minor version of each dependency.
| Component | Deployed On | Supported Version |
| ------------ | --------------------------- | ----------------- |
| Keycloak | NVIDIA Run:ai control plane | 26.6 |
| PostgreSQL | NVIDIA Run:ai control plane | 16 |
| Traefik | NVIDIA Run:ai control plane | 3.7 |
| NATS | NVIDIA Run:ai control plane | 2.11.8 |
| Valkey | NVIDIA Run:ai cluster | 9.0.4 |
| oauth2-proxy | NVIDIA Run:ai cluster | 7.15.4 |
| git-sync | NVIDIA Run:ai cluster | 4.7.0 |
## Supported NVIDIA GPUs
NVIDIA Run:ai multi-tenant v3.0 is compatible with all Data Center GPUs supported by the NVIDIA GPU Operator. Hardware compatibility is determined by the specific version of the GPU Operator deployed within each tenant cluster.
* Supported operator range - NVIDIA Run:ai multi-tenant v3.0 supports GPU Operator versions 25.10 through 26.7.1.
* Hardware verification - To confirm if a specific GPU model is supported, please cross-reference your Operator version with the [Supported NVIDIA Data Center GPUs and Systems](https://docs.nvidia.com/datacenter/cloud-native/gpu-operator/latest/platform-support.html#supported-nvidia-data-center-gpus-and-systems) documentation.
**Note**
* NVIDIA DGX Spark, NVIDIA Jetson and workstations are not supported.
* vGPU is not supported. NVIDIA Run:ai currently supports GPU passthrough only.
## NVIDIA Run:ai Compatible Distributions
NVIDIA Run:ai multi-tenant v3.0 supports a wide range of Kubernetes distributions across on-premises, hybrid, and public cloud environments. Use the table below to verify version requirements for your specific platform.
**Note**
Red Hat OpenShift applies to tenant clusters. The NVIDIA Run:ai control plane runs on Kubernetes.
| Orchestration Platform | Versions | Engine | x86 | ARM |
| -------------------------------------- | -------------------------------- | ---------- | --------- | --------- |
| Upstream Kubernetes | 1.35 - 1.37 | Containerd | Supported | Supported |
| Red Hat OpenShift | 4.20 - 4.22 | CRI-O | Supported | Supported |
| Amazon Elastic Kubernetes Service (EKS) | Based on the upstream Kubernetes | Containerd | Supported | Supported |
| Google Kubernetes Engine (GKE) | Based on the upstream Kubernetes | Containerd | Supported | Supported |
| Azure Kubernetes Service (AKS) | Based on the upstream Kubernetes | Containerd | Supported | Supported |
| Oracle Kubernetes Engine (OKE) | Based on the upstream Kubernetes | Containerd | Supported | Supported |
| Rancher Kubernetes Engine 2 (RKE2) | Based on the upstream Kubernetes | Containerd | Supported | Supported |