Release Notes#
These release notes describe NVIDIA AI Enterprise Infrastructure Release 4.10. Use them to identify the supported infrastructure software components and versions in this release, review compatibility and support information, and locate the per-component release notes for new features, fixed issues, and known limitations.
Note
NVIDIA AI Enterprise Infra 4.10 is a Long-Term Support Branch (LTSB), providing extended support for stable production deployments. For branch lifecycle, support windows, and cross-branch migration guidance (including transition to the next LTSB), refer to the NVIDIA AI Enterprise Lifecycle Policy.
For deployment guidance, refer to the Quick Start Guide. For a full list of supported platforms, hypervisors, operating systems, and orchestration software, refer to the Support Matrix or the interactive support matrix linked under Compatibility and Support. If you are upgrading from 4.9, refer to Upgrading from 4.9 to 4.10.
Latest Release Highlights
NVIDIA Data Center GPU Driver 535.309.01 — Maintenance update within the R535 production driver branch (from 535.288.01 in 4.9). Refer to the 535.309.01 release notes for fixes and platform-support details.
NVIDIA Fabric Manager 535.309.01 — Updated from 535.288.01 in 4.9 in lockstep with the GPU driver, keeping the NVSwitch fabric coordinated with the production driver branch for multi-GPU AI workloads.
NVIDIA vGPU Software 16.14 — NVIDIA Virtual GPU Manager and the NVIDIA vGPU for Compute Guest Driver are both updated from 16.13 to 16.14, refreshing the full vGPU stack in a coordinated release. The single 4.9 “vGPU for Compute” component has also been split into separate Virtual GPU Manager and Guest Driver rows in the supported-software table.
NVIDIA DOCA-OFED Driver 3.3.0 — Updated from 3.2.0 in 4.9, advancing the DOCA-OFED driver stack for high-performance InfiniBand and Ethernet networking on NVIDIA BlueField DPUs and ConnectX SuperNICs.
Kubernetes Operator Major Updates — NVIDIA GPU Operator 26.3.1 (from 25.10.1 in 4.9) and NVIDIA Network Operator 26.1.1 (from 25.10.0 in 4.9) are both major-version bumps. NVIDIA Container Toolkit 1.19.0 (from 1.18.1 in 4.9) also updated for GPU-accelerated container-runtime improvements.
NVIDIA Base Command Manager 11.32.1 — Updated from 11.31.0 in 4.9. Refer to the BCM 11.32.1 release notes for cluster-management and provisioning updates.
What is Included in NVIDIA AI Enterprise Infra 4.10#
Complete list of infrastructure components with versions and documentation links:
Product |
Description |
Version |
|---|---|---|
NVIDIA Data Center GPU Driver |
Latest GPU driver with Blackwell, Hopper, Ada Lovelace, and Ampere architecture support |
|
NVIDIA DOCA-OFED Driver for Networking |
High-performance networking for InfiniBand and Ethernet |
|
NVIDIA Fabric Manager |
Manages NVSwitch fabric to enable high-bandwidth, low-latency GPU-to-GPU communication for multi-GPU AI workloads |
|
NVIDIA Virtual GPU Manager |
GPU driver deployed alongside the hypervisor in virtualized environments. Enables multi-tenant GPU sharing, live migration, and monitoring |
|
NVIDIA vGPU for Compute Guest Driver |
GPU driver deployed in the guest VM to enable multiple VMs to have simultaneous, direct access to a single physical GPU. |
|
NVIDIA Container Toolkit |
GPU-accelerated container runtime with enhanced security |
|
NVIDIA GPU Operator |
Automated GPU software lifecycle management for Kubernetes |
|
NVIDIA Network Operator |
Streamlined networking for GPU workloads in Kubernetes |
|
NVIDIA Base Command Manager (BCM) |
Enterprise cluster provisioning, workload orchestration, and lifecycle management |
Upgrading from 4.9 to 4.10#
Use this checklist when upgrading existing 4.9 deployments to 4.10. Run each step in a maintenance window and validate before proceeding.
Step |
Action |
Reference |
|---|---|---|
|
Read the component version delta in the table above and review per-component release notes for breaking changes, deprecations, and feature additions. |
Component release note links above |
|
Confirm hardware, hypervisor (for virtualized deployments), and operating-system compatibility against the 4.10 support matrix. |
|
|
In virtualized environments, snapshot or back up VMs and capture current driver, operator, and licensing configuration before upgrading. Document the rollback path for each component. |
N/A |
|
Depending on your deployment type i.e. bare metal or virtualized, upgrade the NVIDIA Data Center GPU Driver (applies to bare metal deployments), Virtual GPU Manager (only applies to vGPU for Compute deployments), and DOCA Driver. Fabric Manager is included in the NVIDIA AI Enterprise drivers and updates with the GPU Driver package. |
|
|
Upgrade NVIDIA GPU Operator, Network Operator, NIM Operator, DPU Operator (DPF), and NVIDIA Run:ai (self-hosted) following each operator’s documented upgrade path. NVIDIA Run:ai SaaS upgrades are managed by NVIDIA and require no customer-side upgrade. |
Operator release note links above |
|
For virtualized deployments, upgrade NVIDIA Datacenter driver (GPU passthrough), NVIDIA vGPU for Compute Guest Driver (vGPU deployments), and/or NVIDIA Container Toolkit in tenant VMs. |
|
|
Confirm each licensed vGPU VM reaches the NVIDIA License System and shows Licensed status. |
|
|
Run a representative CUDA or AI/ML workload to confirm performance, feature parity, and operator-managed scheduling behavior. |
N/A |
For lifecycle policy, branch support windows, and migration windows beyond 4.x LTSB (EOL July 2026), refer to the NVIDIA AI Enterprise Lifecycle Policy.
Compatibility and Support#
Support Matrix
Use the Interactive Support Matrix to compare NVIDIA AI Enterprise infrastructure compatibility across releases 4.4 through 4.10. The web tool lets you:
Filter by deployment type, operating system, hypervisor, or orchestration platform.
Search by GPU architecture, platform, Kubernetes distribution, or cloud provider.
Inspect per-configuration release badges and footnote tooltips.
For deep linking, printing, or offline reference, the same information is also available as the static Support Matrix. Both forms cover supported GPU architectures, operating systems, hypervisor and orchestration platform versions, cloud provider instance types, and networking hardware.
Lifecycle and Compatibility Explorer
Use the Interactive Lifecycle and Compatibility Explorer on the NVIDIA AI Enterprise Lifecycle Policy documentation to:
Query by Infrastructure Branch, by release, or by component type and version.
Run a full-stack check from a GPU driver version to validate compatibility and plan upgrades.