Is this page helpful?

NVIDIA AI Enterprise Documentation#

This documentation covers the NVIDIA AI Enterprise Infrastructure Layer software - GPU and network drivers, Kubernetes operators, NVIDIA vGPU for Compute, and NVIDIA Run:ai for AI workload management.

For application-layer software (NIM, NeMo, Omniverse, domain SDKs), refer to Application Software. For enterprise support services, refer to Support.

Quick Start#

🆕 What is New in NVIDIA AI Enterprise Infra 7.7#

Latest Release Highlights

  • NVIDIA Data Center GPU Driver and Fabric Manager - Maintenance update within the R580 production driver branch (from 580.167.08 in 7.6). NVIDIA Data Center GPU Driver and NVIDIA Fabric Manager update to 580.173.02 in lockstep. Refer to the 580.173.02 release notes for fixes and platform-support details.

  • NVIDIA vGPU for Compute Updates - The Linux Guest Driver updates to 580.173.02 (from 580.167.08 in 7.6). The Windows Guest Driver updates to 582.70 (from 582.53). NVIDIA Virtual GPU Manager remains at 580.159.01.

View Full 7.7 Release Notes

Previous Releases#

📋 Release 7.6 Highlights

  • Maintenance update within the R580 production driver branch (from 580.159.03 in 7.5). NVIDIA Fabric Manager also updates to 580.167.08 in lockstep with the driver. Refer to the 580.167.08 release notes for fixes and platform-support details.

  • NVIDIA vGPU for Compute Updates - The Linux Guest Driver updates to 580.167.08 (from 580.159.03 in 7.5). The Windows Guest Driver remains at 582.53. NVIDIA Virtual GPU Manager remains at 580.159.01.

  • Software Component Updates - NVIDIA DOCA Driver for Networking and DOCA Microservices update to 3.4.0 (from 3.3.0), NVIDIA DPU Operator (DPF) to 26.4.0 (from 25.10.1), NVIDIA Network Operator to 26.4.0 (from 26.1.1), NVIDIA GPU Operator to 26.3.3 (from 26.3.1), NVIDIA Container Toolkit to 1.19.1 (from 1.19.0), NVIDIA NIM Operator to 3.1.1 (from 3.1.0), and NVIDIA Base Command Manager to 11.33.1 (from 11.32.1).

  • Partner-Led Support - NVIDIA AI Enterprise is also available through SUSE AI Factory with NVIDIA, where SUSE delivers technical support directly for deployments on SLES. Refer to Partner-Led Support in the Support Matrix for details.

View Full 7.6 Release Notes

📦 Archived Releases (7.0–7.5)

Earlier NVIDIA AI Enterprise 7.x releases with key component versions:

  • 7.5 Release Notes - Maintenance update advancing the NVIDIA Data Center GPU Driver to 580.159.03 (Linux vGPU for Compute Guest Driver in lockstep, Windows 582.53), with NVIDIA Virtual GPU Manager 580.159.01, DOCA Driver 3.3.0, and Base Command Manager 11.32.1

  • 7.4 Release Notes - NVIDIA Data Center GPU Driver 580.126.09 (Blackwell architecture support), NVIDIA Virtual GPU Manager 580.126.08 and vGPU for Compute Guest Driver (Linux 580.126.09, Windows 582.16), DOCA Driver 3.2.0, Base Command Manager 11.31.0

  • 7.3 Release Notes - NVIDIA Data Center GPU Driver 580.105.08 (adds DGX B300, HGX B300, and GB300 NVL72 support), NVIDIA Virtual GPU Manager 580.105.06 and vGPU for Compute Guest Driver (Linux 580.105.08, Windows 581.80), DOCA-OFED 3.1.0, Base Command Manager 11.25.08

  • 7.2 Release Notes - NVIDIA Data Center GPU Driver 580.95.05, NVIDIA Virtual GPU Manager 580.95.02 and vGPU for Compute Guest Driver (Linux 580.95.05, Windows 581.42) (RTX PRO 6000 Blackwell Server Edition support on VMware vSphere), DOCA-OFED 3.1.0, Base Command Manager 11.25.08

  • 7.1 Release Notes - NVIDIA Data Center GPU Driver 580.82.07, NVIDIA Virtual GPU Manager 580.82.02 and vGPU for Compute Guest Driver (Linux 580.82.07, Windows 581.15), DOCA-OFED 3.1.0, Base Command Manager 11.25.05

  • 7.0 Release Notes - Initial R580 driver branch (580.65.06) and NVIDIA Virtual GPU Manager 580.65.05 and vGPU for Compute Guest Driver (Linux 580.65.06, Windows 580.88) with HGX B200 and RTX PRO 6000 Blackwell Server Edition support, DOCA-OFED 3.0.0, Base Command Manager 11.25.05