{"id":"nvidia-inference-context-memory-storage-platform--nvidianews-nvidia-com","name":"NVIDIA Inference Context Memory Storage Platform (now NVIDIA CMX Context Memory Storage Platform)","url":"https://www.nvidia.com/en-us/data-center/ai-storage/cmx/","page":"https://mrpeppersdev.github.io/agent-infrastructure-landscape/systems/nvidia-inference-context-memory-storage-platform--nvidianews-nvidia-com","family":"memory","section":"Dedicated memory layers","sections":["Dedicated memory layers"],"tier":2,"last_verified_at":"2026-10-23","lifecycle":{"status":"renamed","severity":1,"note":"now NVIDIA CMX Context Memory Storage Platform"},"taxonomy":{"storage":[{"value":"kv","primary":true,"reason":null}],"retrieval":[{"value":"similarity","primary":true,"reason":null}],"persistence":[{"value":"session","primary":true,"reason":null}],"update":[{"value":"replacement","primary":true,"reason":null}],"unit":[{"value":"token","primary":true,"reason":null}],"governance":[{"value":"opaque","primary":true,"reason":null}],"conflict":[{"value":"n/a","primary":true,"reason":"not applicable — KV-cache layer (no semantic conflict resolution)"}]},"cells":{"type":{"value":"Hardware-accelerated cluster KV-cache storage tier","citation":"https://nvidianews.nvidia.com/news/nvidia-bluefield-4-powers-new-class-of-ai-native-storage-infrastructure-for-the-next-frontier-of-ai","status":"real-data","tier":"T2","source_class":"vendor"},"desc":{"value":"Renamed NVIDIA CMX Context Memory Storage Platform (2026). A BlueField-4 (STX) powered \"AI-native context tier\" that extends GPU memory with a shared pod-level KV-cache tier for long-context, multi-turn agentic inference, with Dynamo routing and DOCA Memos SDK managing cache across compute and storage nodes. Announced CES Jan 2026; offered via storage partners (Dell, HPE, IBM, NetApp, VAST, WEKA, etc.).","citation":"https://www.nvidia.com/en-us/data-center/ai-storage/cmx/","status":"real-data","tier":"T2","source_class":"vendor"},"claims":{"value":"Up to 5x tokens/sec; up to 5x greater power efficiency vs traditional storage; reduced time-to-first-token in multi-turn","citation":"https://nvidianews.nvidia.com/news/nvidia-bluefield-4-powers-new-class-of-ai-native-storage-infrastructure-for-the-next-frontier-of-ai","status":"real-data","tier":"T2","source_class":"vendor"},"modalities":{"value":"text (KV-cache; modality-agnostic at storage layer)","citation":"https://nvidianews.nvidia.com/news/nvidia-bluefield-4-powers-new-class-of-ai-native-storage-infrastructure-for-the-next-frontier-of-ai","status":"real-data","tier":"T2","source_class":"vendor"},"created":{"value":"Announced 2026-04 (NVIDIA GTC season); GA H2 2026 (BlueField-4 GA dependency)","citation":"https://nvidianews.nvidia.com/news/nvidia-bluefield-4-powers-new-class-of-ai-native-storage-infrastructure-for-the-next-frontier-of-ai","status":"real-data","tier":"T2","last_verified_at":"2026-05-12","source_class":"vendor"},"latest-release":{"value":"not applicable — hardware platform; not OSS","citation":null,"status":"not-applicable","tier":"T3","last_verified_at":"2026-05-12"},"license":{"value":"proprietary (hardware/DOCA) + Dynamo Apache-2.0","citation":"https://github.com/ai-dynamo/dynamo","status":"real-data","tier":"T1","source_class":"repo"},"gh":{"value":"not applicable — proprietary platform","citation":null,"status":"not-applicable","tier":"T3","last_verified_at":"2026-05-12"},"mindshare":{"value":"NVIDIA newsroom + The Register + HPCwire + StorageReview coverage 2026-04","citation":"https://nvidianews.nvidia.com/news/nvidia-bluefield-4-powers-new-class-of-ai-native-storage-infrastructure-for-the-next-frontier-of-ai","status":"real-data","tier":"T2","last_verified_at":"2026-05-12","source_class":"vendor"},"citations":{"value":"not applicable — not academic","citation":null,"status":"not-applicable","tier":"T3","last_verified_at":"2026-05-12"},"funding":{"value":"Public (NVDA); ~$3T+ market cap (2026)","citation":"https://www.nvidia.com/","status":"real-data","tier":"T2","last_verified_at":"2026-05-12","source_class":"vendor"},"customers":{"value":"Design-win partners: AIC, Cloudian, DDN, Dell Technologies, HPE, Hitachi Vantara, IBM, Nutanix, Pure Storage, Supermicro, VAST Data, WEKA","citation":"https://nvidianews.nvidia.com/news/nvidia-bluefield-4-powers-new-class-of-ai-native-storage-infrastructure-for-the-next-frontier-of-ai","status":"real-data","tier":"T2","source_class":"vendor"},"pricing":{"value":"searched not found","citation":"https://nvidianews.nvidia.com/news/nvidia-bluefield-4-powers-new-class-of-ai-native-storage-infrastructure-for-the-next-frontier-of-ai","status":"depth-floor-reached","tier":"T2","source_class":"vendor"},"compliance":{"value":"SOC 2 / ISO via partner storage vendor (NVIDIA stack itself enterprise-grade)","citation":"https://www.nvidia.com/en-us/security/","status":"real-data","tier":"T2","source_class":"vendor"},"data-handling":{"value":"RDMA-only data path; tenant isolation handled by partner storage controllers","citation":"https://www.nvidia.com/en-us/data-center/products/bluefield/","status":"real-data","tier":"T2","source_class":"vendor"},"deployment":{"value":"On-prem rack-scale (NVIDIA Rubin cluster + partner storage); managed via DOCA","citation":"https://nvidianews.nvidia.com/news/nvidia-bluefield-4-powers-new-class-of-ai-native-storage-infrastructure-for-the-next-frontier-of-ai","status":"real-data","tier":"T2","source_class":"vendor"},"hq":{"value":"US (Santa Clara CA)","citation":"https://www.nvidia.com/","status":"real-data","tier":"T2","source_class":"vendor"},"founders":{"value":"NVIDIA; co-founders Jensen Huang, Chris Malachowsky, Curtis Priem (1993)","citation":"https://en.wikipedia.org/wiki/Nvidia","status":"real-data","tier":"T2","source_class":"third-party"},"volume":{"value":"Petabyte-scale KV-cache capacity (cluster-level)","citation":"https://nvidianews.nvidia.com/news/nvidia-bluefield-4-powers-new-class-of-ai-native-storage-infrastructure-for-the-next-frontier-of-ai","status":"real-data","tier":"T2","source_class":"vendor"},"perf":{"value":"5x tokens/sec and 5x power efficiency multipliers (no absolute baseline disclosed)","citation":"https://nvidianews.nvidia.com/news/nvidia-bluefield-4-powers-new-class-of-ai-native-storage-infrastructure-for-the-next-frontier-of-ai","status":"real-data","tier":"T2","source_class":"vendor"},"repro":{"value":"not applicable — not a research paper","citation":null,"status":"not-applicable","tier":"T3"},"code-release":{"value":"Code public (Dynamo/NIXL OSS) + hardware proprietary","citation":"https://github.com/ai-dynamo/dynamo","status":"real-data","tier":"T1","source_class":"repo"},"api-surface":{"value":"DOCA APIs; NIXL transfer library; Dynamo runtime","citation":"https://developer.nvidia.com/dynamo","status":"real-data","tier":"T2","source_class":"vendor"},"latency":{"value":"sub-microsecond RDMA fabric (Spectrum-X)","citation":"https://www.nvidia.com/en-us/networking/spectrumx/","status":"real-data","tier":"T2","source_class":"vendor"},"throughput":{"value":"line-rate 800G Ethernet (Spectrum-X)","citation":"https://www.nvidia.com/en-us/networking/spectrumx/","status":"real-data","tier":"T2","source_class":"vendor"},"backend-storage":{"value":"custom (BlueField-4 + partner SSD/NVMe arrays)","citation":"https://nvidianews.nvidia.com/news/nvidia-bluefield-4-powers-new-class-of-ai-native-storage-infrastructure-for-the-next-frontier-of-ai","status":"real-data","tier":"T2","source_class":"vendor"},"multi-tenancy":{"value":"hard-isolation (BlueField DPU enforces)","citation":"https://www.nvidia.com/en-us/data-center/products/bluefield/","status":"real-data","tier":"T2","source_class":"vendor"},"encryption":{"value":"at-rest + in-transit (DPU-accelerated crypto)","citation":"https://www.nvidia.com/en-us/data-center/products/bluefield/","status":"real-data","tier":"T2","source_class":"vendor"},"sso-rbac":{"value":"partner-dependent (storage vendor handles SSO/RBAC)","citation":"https://nvidianews.nvidia.com/news/nvidia-bluefield-4-powers-new-class-of-ai-native-storage-infrastructure-for-the-next-frontier-of-ai","status":"real-data","tier":"T2","source_class":"vendor"},"embedding-model":{"value":"not applicable — KV cache (raw tokens; no embedding model)","citation":null,"status":"not-applicable","tier":"T3"},"consistency":{"value":"strong (RDMA in-cluster)","citation":"https://www.nvidia.com/en-us/networking/spectrumx/","status":"real-data","tier":"T2","source_class":"vendor"},"versioning":{"value":"searched not found","citation":"https://nvidianews.nvidia.com/news/nvidia-bluefield-4-powers-new-class-of-ai-native-storage-infrastructure-for-the-next-frontier-of-ai","status":"depth-floor-reached","tier":"T2","source_class":"vendor"},"tombstoning":{"value":"replacement (LRU/LFU at KV-cache layer)","citation":"https://developer.nvidia.com/dynamo","status":"real-data","tier":"T2","source_class":"vendor"},"schema-evolution":{"value":"not applicable — KV cache schema fixed (token-tensor)","citation":null,"status":"not-applicable","tier":"T3"},"namespace":{"value":"per-cluster / per-tenant (DPU-enforced)","citation":"https://www.nvidia.com/en-us/data-center/products/bluefield/","status":"real-data","tier":"T2","source_class":"vendor"},"contradiction":{"value":"not applicable — KV cache (no semantic contradiction)","citation":null,"status":"not-applicable","tier":"T3"},"forgetting":{"value":"LRU/replacement (cache eviction)","citation":"https://developer.nvidia.com/dynamo","status":"real-data","tier":"T2","source_class":"vendor"},"mcp-support":{"value":"searched not found","citation":"https://nvidianews.nvidia.com/news/nvidia-bluefield-4-powers-new-class-of-ai-native-storage-infrastructure-for-the-next-frontier-of-ai","status":"depth-floor-reached","tier":"T2","source_class":"vendor"},"a2a-support":{"value":"searched not found","citation":"https://nvidianews.nvidia.com/news/nvidia-bluefield-4-powers-new-class-of-ai-native-storage-infrastructure-for-the-next-frontier-of-ai","status":"depth-floor-reached","tier":"T2","source_class":"vendor"},"otel":{"value":"first-class (DOCA telemetry / OTel via partner integrations)","citation":"https://developer.nvidia.com/networking/doca","status":"real-data","tier":"T2","source_class":"vendor"},"webhooks":{"value":"searched not found","citation":"https://nvidianews.nvidia.com/news/nvidia-bluefield-4-powers-new-class-of-ai-native-storage-infrastructure-for-the-next-frontier-of-ai","status":"depth-floor-reached","tier":"T2","source_class":"vendor"},"import-export":{"value":"partner-dependent (storage vendor APIs for export)","citation":"https://nvidianews.nvidia.com/news/nvidia-bluefield-4-powers-new-class-of-ai-native-storage-infrastructure-for-the-next-frontier-of-ai","status":"real-data","tier":"T2","source_class":"vendor"},"integration-count":{"value":"not applicable — wrong section","citation":null,"status":"not-applicable","tier":"T3"},"orchestration":{"value":"not applicable — KV-cache storage tier (no agent orchestration model)","citation":null,"status":"not-applicable","tier":"T3"},"programmatic-control":{"value":"not applicable — wrong section","citation":null,"status":"not-applicable","tier":"T3"},"vendor-benchmarks":{"value":"not applicable — wrong section","citation":null,"status":"not-applicable","tier":"T3","last_verified_at":"2026-05-12"},"pricing-specifics":{"value":"not applicable — wrong section","citation":null,"status":"not-applicable","tier":"T3"},"agent-abstraction":{"value":"not applicable — wrong section","citation":null,"status":"not-applicable","tier":"T3"},"memory-primitives":{"value":"not applicable — wrong section","citation":null,"status":"not-applicable","tier":"T3"},"llm-lock":{"value":"not applicable — wrong section","citation":null,"status":"not-applicable","tier":"T3"},"runtimes":{"value":"not applicable — wrong section","citation":null,"status":"not-applicable","tier":"T3"},"session-handling":{"value":"not applicable — wrong section","citation":null,"status":"not-applicable","tier":"T3"},"validated-verticals":{"value":"gigascale agentic inference operators; cloud GPU/HPC providers; LLM inference vendors (design-win partners: AIC/Cloudian/DDN/Dell/HPE/Hitachi Vantara/IBM/Nutanix/Pure Storage/Supermicro/VAST Data/WEKA)","citation":"https://nvidianews.nvidia.com/news/nvidia-bluefield-4-powers-new-class-of-ai-native-storage-infrastructure-for-the-next-frontier-of-ai","status":"real-data","tier":"T2","source_class":"vendor"},"time-to-running":{"value":"months (rack-scale BlueField-4 deployment; depends on Rubin GA H2 2026)","citation":"https://nvidianews.nvidia.com/news/nvidia-bluefield-4-powers-new-class-of-ai-native-storage-infrastructure-for-the-next-frontier-of-ai","status":"real-data","tier":"T2","source_class":"vendor"},"anti-fit":{"value":"not for non-NVIDIA-Rubin/BlueField stacks; not for small-deployment / edge use; not for non-KV-cache memory needs (semantic conflict resolution etc.)","citation":"https://nvidianews.nvidia.com/news/nvidia-bluefield-4-powers-new-class-of-ai-native-storage-infrastructure-for-the-next-frontier-of-ai","status":"real-data","tier":"T2","source_class":"vendor"},"optimised-for":{"value":"petabyte-scale RDMA-accelerated KV-cache placement (5x tokens/sec; 5x power efficiency); multi-turn TTFT reduction","citation":"https://nvidianews.nvidia.com/news/nvidia-bluefield-4-powers-new-class-of-ai-native-storage-infrastructure-for-the-next-frontier-of-ai","status":"real-data","tier":"T2","source_class":"vendor"},"adjacent-infrastructure":{"value":"requires NVIDIA Rubin cluster + BlueField-4 DPU + Spectrum-X Ethernet + DOCA + NIXL + Dynamo; partner storage (12 named OEMs)","citation":"https://nvidianews.nvidia.com/news/nvidia-bluefield-4-powers-new-class-of-ai-native-storage-infrastructure-for-the-next-frontier-of-ai","status":"real-data","tier":"T2","source_class":"vendor"},"pros":{"value":"Bridges GPU HBM and storage with hardware-accelerated context offload; broad partner ecosystem; aligned with Rubin cluster architecture; positions KV-cache as a first-class storage tier.","citation":"https://nvidianews.nvidia.com/news/nvidia-bluefield-4-powers-new-class-of-ai-native-storage-infrastructure-for-the-next-frontier-of-ai","status":"real-data","tier":"T2","source_class":"vendor"},"cons":{"value":"GA H2 2026 — not yet shipping; locked to NVIDIA Rubin/BlueField stack; multipliers reported without baseline absolute numbers; opaque tenant-isolation details.","citation":"https://nvidianews.nvidia.com/news/nvidia-bluefield-4-powers-new-class-of-ai-native-storage-infrastructure-for-the-next-frontier-of-ai","status":"real-data","tier":"T2","source_class":"vendor"},"links":{"value":"announcement; Dynamo OSS; DOCA platform; BlueField page","citation":"https://nvidianews.nvidia.com/news/nvidia-bluefield-4-powers-new-class-of-ai-native-storage-infrastructure-for-the-next-frontier-of-ai","status":"real-data","tier":"T2","source_class":"vendor"},"obs-langsmith":{"value":"","citation":null,"status":"no-data","tier":"T3","last_verified_at":"2026-05-14"},"obs-opentelemetry":{"value":"","citation":null,"status":"no-data","tier":"T3","last_verified_at":"2026-05-14"},"obs-datadog":{"value":"","citation":null,"status":"no-data","tier":"T3","last_verified_at":"2026-05-14"},"obs-helicone":{"value":"","citation":null,"status":"no-data","tier":"T3","last_verified_at":"2026-05-14"},"obs-weave":{"value":"","citation":null,"status":"no-data","tier":"T3","last_verified_at":"2026-05-14"},"obs-langfuse":{"value":"","citation":null,"status":"no-data","tier":"T3","last_verified_at":"2026-05-14"},"obs-arize":{"value":"","citation":null,"status":"no-data","tier":"T3","last_verified_at":"2026-05-14"},"obs-custom":{"value":"","citation":null,"status":"no-data","tier":"T3","last_verified_at":"2026-05-14"},"cost-token-budget":{"value":"","citation":null,"status":"no-data","tier":"T3","last_verified_at":"2026-05-14"},"cost-prompt-caching":{"value":"","citation":null,"status":"no-data","tier":"T3","last_verified_at":"2026-05-14"},"cost-semantic-caching":{"value":"","citation":null,"status":"no-data","tier":"T3","last_verified_at":"2026-05-14"},"cost-batching":{"value":"","citation":null,"status":"no-data","tier":"T3","last_verified_at":"2026-05-14"},"cost-model-routing":{"value":"","citation":null,"status":"no-data","tier":"T3","last_verified_at":"2026-05-14"},"cost-streaming-only":{"value":"","citation":null,"status":"no-data","tier":"T3","last_verified_at":"2026-05-14"},"cost-observability-cost-attribution":{"value":"","citation":null,"status":"no-data","tier":"T3","last_verified_at":"2026-05-14"},"eval-langsmith-evals":{"value":"","citation":null,"status":"no-data","tier":"T3","last_verified_at":"2026-05-14"},"eval-braintrust":{"value":"","citation":null,"status":"no-data","tier":"T3","last_verified_at":"2026-05-14"},"eval-weights-and-biases-agent":{"value":"","citation":null,"status":"no-data","tier":"T3","last_verified_at":"2026-05-14"},"eval-helicone-evals":{"value":"","citation":null,"status":"no-data","tier":"T3","last_verified_at":"2026-05-14"},"eval-custom-test-harness":{"value":"","citation":null,"status":"no-data","tier":"T3","last_verified_at":"2026-05-14"},"eval-human-loop":{"value":"","citation":null,"status":"no-data","tier":"T3","last_verified_at":"2026-05-14"},"eval-production-traffic-replay":{"value":"","citation":null,"status":"no-data","tier":"T3","last_verified_at":"2026-05-14"},"commit-trajectory":{"value":"not applicable — no GitHub repo","citation":null,"status":"not-applicable","tier":"T3","last_verified_at":"2026-05-14"},"citation-trajectory":{"value":"not applicable — non-paper row","citation":null,"status":"not-applicable","tier":"T3","last_verified_at":"2026-05-14"},"download-trajectory":{"value":"not applicable — not published as a package","citation":null,"status":"not-applicable","tier":"T3","last_verified_at":"2026-05-14"},"cost-input-usd-per-mtok":{"value":"searched not found","citation":"https://nvidianews.nvidia.com/news/nvidia-bluefield-4-powers-new-class-of-ai-native-storage-infrastructure-for-the-next-frontier-of-ai","status":"depth-floor-reached","tier":"T2","source_class":"vendor"},"cost-output-usd-per-mtok":{"value":"searched not found","citation":"https://nvidianews.nvidia.com/news/nvidia-bluefield-4-powers-new-class-of-ai-native-storage-infrastructure-for-the-next-frontier-of-ai","status":"depth-floor-reached","tier":"T2","source_class":"vendor"},"cost-tier":{"value":"searched not found","citation":"https://nvidianews.nvidia.com/news/nvidia-bluefield-4-powers-new-class-of-ai-native-storage-infrastructure-for-the-next-frontier-of-ai","status":"depth-floor-reached","tier":"T2","source_class":"vendor"},"cost-pricing-model":{"value":"searched not found","citation":"https://nvidianews.nvidia.com/news/nvidia-bluefield-4-powers-new-class-of-ai-native-storage-infrastructure-for-the-next-frontier-of-ai","status":"depth-floor-reached","tier":"T2","source_class":"vendor"},"cost-last-verified":{"value":"searched not found","citation":"https://nvidianews.nvidia.com/news/nvidia-bluefield-4-powers-new-class-of-ai-native-storage-infrastructure-for-the-next-frontier-of-ai","status":"depth-floor-reached","tier":"T2","source_class":"vendor"},"capability-composite-score":{"value":"not applicable — hardware-accelerated KV-cache storage tier, not a capable inferencing system itself","citation":null,"status":"not-applicable","tier":"T3"},"capability-band":{"value":"not applicable — hardware-accelerated KV-cache storage tier, not a capable inferencing system itself","citation":null,"status":"not-applicable","tier":"T3"},"capability-benchmark-sources":{"value":"not applicable — hardware-accelerated KV-cache storage tier, not a capable inferencing system itself","citation":null,"status":"not-applicable","tier":"T3"},"capability-last-verified":{"value":"not applicable — hardware-accelerated KV-cache storage tier, not a capable inferencing system itself","citation":null,"status":"not-applicable","tier":"T3"},"use-case-tags":{"value":"long-running-session, multi-agent-coordination, analytical-summarization","citation":null,"status":"estimate","tier":"T3"},"use-case-anti-tags":{"value":"offline-capable, scoped-agentic","citation":null,"status":"estimate","tier":"T3"},"adoption-accelerator":{"value":"","citation":null,"status":"no-data","tier":"T3"},"architecture-deployment-locus":{"value":"local","citation":"https://www.nvidia.com/en-us/data-center/ai-storage/cmx/","status":"real-data","tier":"T2","source_class":"vendor"},"licensing-bucket":{"value":"commercial","citation":"https://github.com/ai-dynamo/dynamo","status":"real-data","tier":"T1","source_class":"repo"},"positioning-platform-vs-product":{"value":"platform","citation":"https://www.nvidia.com/en-us/data-center/ai-storage/cmx/","status":"real-data","tier":"T2","source_class":"vendor"},"taxonomy-three-layer-rollup":{"value":"agent-infra","citation":"https://github.com/MrPeppersDev/agent-infrastructure-landscape/blob/main/docs/schema/section-to-three-layer-rollup.json","status":"real-data","tier":"T1","source_class":"repo"},"developer-activity-opendigger":{"value":"","citation":null,"status":"no-data","tier":"T3"},"freshness-date":{"value":"2026-10-23","citation":"https://mrpeppersdev.github.io/agent-infrastructure-landscape/systems/nvidia-inference-context-memory-storage-platform--nvidianews-nvidia-com","status":"real-data","tier":"T2","source_class":"catalog"},"compliance-certifications":{"value":"","citation":null,"status":"no-data","tier":"T3"},"sdk-languages":{"value":"","citation":null,"status":"no-data","tier":"T3"},"protocol-support":{"value":"","citation":null,"status":"no-data","tier":"T3"},"maturity":{"value":"","citation":null,"status":"no-data","tier":"T3"}},"corroboration":{"cited_cells":42,"third_party_cells":1},"edges":{"outgoing":[],"incoming":[]},"history":[{"date":"2026-10-23","commit":"5fd0959f49","pr":223,"kind":"lifecycle","changes":[{"field":"name","label":"Name","from":"NVIDIA Inference Context Memory Storage Platform","to":"NVIDIA Inference Context Memory Storage Platform (now NVIDIA CMX Context Memory Storage Platform)"},{"field":"url","label":"Homepage","from":"https://nvidianews.nvidia.com/news/nvidia-bluefield-4-powers-new-class-of-ai-native-storage-infrastructure-for-the-next-frontier-of-ai","to":"https://www.nvidia.com/en-us/data-center/ai-storage/cmx/"}]},{"date":"2026-10-22","commit":"d1f77cf661","pr":219,"kind":"update","changes":[{"field":"license","label":"License","from":"proprietary (BlueField-4 firmware) + NIXL/Dynamo software stack (mixed)","to":"proprietary (hardware/DOCA) + Dynamo Apache-2.0","source":"https://github.com/ai-dynamo/dynamo"}]}],"license":"CC-BY-4.0","source":"https://github.com/MrPeppersDev/agent-infrastructure-landscape/blob/main/data/landscape.json"}