{"resourceId":"nvidia-aie-82-support-matrix-2026","versions":[{"version":"external-065051557cb29b36d65bba0d4456f1ee4fc2aecd8caa7f1eb3df9235436033be","resource":{"id":"nvidia-aie-82-support-matrix-2026","title":"AI Enterprise Infra 8.2 support matrix pins operators and Government Ready columns","organization":"NVIDIA","sector":"Enterprise AI infrastructure software","geography":"Global (enterprise/gov packaging)","publishedAt":"undated 8.2 matrix page (companion to Infra 8.2)","publicationDate":null,"eventDate":null,"sourceName":"NVIDIA AI Enterprise Documentation","sourceLabel":"Vendor support matrix","sourceUrl":"https://docs.nvidia.com/ai-enterprise/release-8/latest/support/support-matrix-8/8.2.html","evidenceClass":"vendor-claim","outcomeClass":"effective","topics":["infrastructure","governance-procurement","data-security"],"finding":"The version-pinned AI Enterprise Infra 8.2 support matrix lists component versions and a Government Ready column: Data Center GPU Driver 595.91.07, Fabric Manager 595.91.07, GPU Operator 26.3.3 (Supported for Gov Ready with footnote to operand list; ARM N/A for Gov Ready), Network Operator 26.4.1, NIM Operator 3.1.2 (Supported across x86/ARM/Gov Ready), DOCA networking 3.4.0 Supported for Gov Ready, Run:ai 2.26, BCM 11.33.1. Platform tables include HGX/DGX B200/B300 and GB200/GB300 NVL72 bare-metal notes, OpenShift 4.18–4.22, Kubernetes 1.32–1.36, and cloud instance matrices. Footnotes restrict some HGX vGPU/KubeVirt combinations.","sledRelevance":"Actionable for SLED regulated Kubernetes baselines: yesterday 8.2 release notes are complemented by the matrix Government Ready flags and orchestration/OS pins needed for ATO-oriented stacks. Does not itself confer FedRAMP authorization.","evidence":"Table 4 infrastructure software with Gov Ready column; Kubernetes bare-metal and virtualized tables; GB200/GB300 bare-metal-only notes; footnotes pointing to Gov Ready operand docs; vSphere HGX B200/B300 1:1 and multi-vGPU only (no fractional).","architectureImplications":"Pin operator/driver versions to 8.2 matrix before claiming support; GB300 NVL72 paths are bare-metal scoped in matrix; Gov Ready ARM gaps matter for Grace hosts.","governanceImplications":"Use matrix as compliance inventory input; still need host FIPS/STIG and system ATO controls outside this page.","securityPrivacyImplications":"Gov Ready column is packaging designation; not a completed authorization.","caveats":"Vendor matrix; undated page body; Gov Ready footnotes defer to other docs; some cells N/A.","streamIds":["nvidia"],"roles":{"sales":"For regulated buyers, bring the 8.2 matrix—not just marketing Government Ready. Show which operators carry the designation (GPU/NIM/Network) and which driver cells are N/A, then propose a matrix-conformant BOM review. Do not imply the matrix equals an ATO or FedRAMP authorization package. Keep claims bounded to the inspected evidence and require site-matched proof before any capacity or interconnect commitment. Keep claims bounded to the inspected evidence and require site-matched proof before any capacity or interconnect commitment.","engineering":"Fit as the compatibility source of truth for Infra 8.2 upgrades. Integrate pinned GPU Operator 26.3.3, NIM Operator 3.1.2, driver 595.91.07, and validated K8s/OpenShift ranges. Proof of value: cluster inventory matches matrix before workload cutover on regulated hosts. Document configuration pins and failure modes so the next upgrade cannot silently invalidate the measured result. Document configuration pins and failure modes so the next upgrade cannot silently invalidate the measured result.","delivery":"Work is upgrade planning: diff current pins, stage operator upgrades, validate Gov Ready operands where required. Ownership: platform engineering. Risks: fractional vGPU assumptions on HGX B200/B300 vSphere; ARM Gov Ready gaps; mistaking matrix support for system authorization. Record acceptance evidence in the program archive so later editions can cite what was actually proven. Record acceptance evidence in the program archive so later editions can cite what was actually proven."},"roleDetails":{"sales":{"customerProblem":"Regulated buyers need version-pinned, Gov Ready-aware BOMs—not vague enterprise bundles.","stakeholders":["platform","security/compliance","procurement"],"discoveryQuestions":["Which matrix row is your target orchestration?","Do you require Gov Ready operands?","Bare-metal NVL72 or virtualized?"],"valueHypothesis":"An 8.2 matrix-conformant stack reduces support and compliance ambiguity for NVIDIA AI Enterprise.","potentialEngagement":"BOM conformance review against 8.2 matrix plus Gov Ready footnotes.","unsupportedClaims":["Matrix equals FedRAMP ATO","All components Gov Ready","Fractional vGPU on HGX B200/B300 vSphere"]},"engineering":{"fit":"Any AI Enterprise Infra 8.2 deployment planning upgrade or greenfield.","architectureIntegration":"Pin drivers/operators/orchestrators per tables; respect bare-metal NVL72 notes.","prerequisites":["current inventory","NGC entitlement","Gov Ready operand list if required"],"constraints":["Footnote-deferred Gov Ready details","HGX virtualization limits","ARM Gov Ready N/A on GPU Operator"],"security":"Treat Gov Ready as necessary but not sufficient for system authorization.","proofOfValue":["Inventory matches pins","Smoke tests on supported OS/K8s","Gov Ready images pull successfully"]},"delivery":{"work":"Version-pin upgrade runbook and validation matrix.","dependencies":["NVIDIA entitlement","OS vendor","security reviewers"],"ownership":"Platform engineering lead.","skillsAdoption":"Ops must read support matrix footnotes, not only release-note headlines.","governanceCheckpoints":["No unsupported orchestration versions","Gov Ready scope documented","vGPU mode restrictions checked"],"acceptanceCriteria":["Pins verified","Workloads healthy","Exception log empty or approved"],"risks":["Silent drift from pins","Virtualization mode mismatch","Authorization assumed"]}},"retrievedAt":"2026-09-20T03:07:30Z","enrichedAt":"2026-09-20T03:13:48Z","enrichmentBasis":"retrieved source","accessibilityWorkforceImplications":"None stated.","procurementImplications":"Write 8.2 matrix pins into contracts and acceptance tests.","operatingModelImplications":"Supportability becomes a continuous pin-management practice.","sourceVerification":{"openedUrl":"https://docs.nvidia.com/ai-enterprise/release-8/latest/support/support-matrix-8/8.2.html","referenceExcerpt":"This page is the version-pinned support matrix for NVIDIA AI Enterprise 8.2.","promptVersion":"sled-research-v4.0","model":null,"basis":"agent-reported inspection"}}}]}