{
  "data_version": "1",
  "schema": "https://whatledto.com/data",
  "slug": "nvidia-datacenter-gpu-roadmap",
  "title": "Nvidia Data Center GPUs: Hopper, Blackwell and Vera Rubin",
  "url": "https://whatledto.com/events/nvidia-datacenter-gpu-roadmap",
  "markdown": "https://whatledto.com/events/nvidia-datacenter-gpu-roadmap.md",
  "feed": "https://whatledto.com/events/nvidia-datacenter-gpu-roadmap/feed.xml",
  "vertical": "tech-industry",
  "topic": "Tech industry",
  "status": "live",
  "tags": [
    {
      "slug": "ai-accelerators",
      "name": "AI accelerators",
      "url": "https://whatledto.com/tags/ai-accelerators"
    },
    {
      "slug": "data-center-gpus",
      "name": "Data center GPUs",
      "url": "https://whatledto.com/tags/data-center-gpus"
    },
    {
      "slug": "nvidia",
      "name": "Nvidia",
      "url": "https://whatledto.com/tags/nvidia"
    }
  ],
  "created_at": "2026-09-02T06:54:49.616441+00:00",
  "updated_at": "2026-09-09T06:05:44.107268+00:00",
  "last_researched_at": "2026-09-09T00:53:43.564444+00:00",
  "tldr": "Nvidia announced the Hopper H100 in March 2022 and said it began shipping Hopper-based H100 GPUs in November 2022. Blackwell was announced in March 2024, then encountered a mask change intended to improve yields; Nvidia later reported billions of dollars of Blackwell sales in its first sales quarter and said Blackwell Ultra production was ramping in August 2025. Nvidia has moved to Vera Rubin: it declared the platform in full production in March 2026, while its August filing said production shipments had begun in fiscal Q3 2027; data center revenue was $89.0 billion in the quarter reported August 26. Rubin Ultra remains scheduled for the second half of 2027, but Nvidia's statement that its roadmap was intact did not clarify a report that the Rubin Ultra Kyber rack had slipped to 2028, and the provided sources give no release timing or specifications for Feynman.",
  "hero_image": "https://images.unsplash.com/photo-1555618565-9f2b0323a10d?ixid=M3w5OTg1Nzd8MHwxfGFsbHx8fHx8fHx8fDE3ODgzNDk0NjB8&ixlib=rb-4.1.0&w=1600&q=80&auto=format&fit=crop",
  "hero_credit": {
    "source": "unsplash",
    "photo_id": "TErYPw4o1KM",
    "photographer": "Christian Wiediger",
    "photographer_url": "https://unsplash.com/@christianw?utm_source=whatledto&utm_medium=referral",
    "photo_url": "https://unsplash.com/photos/a-close-up-of-a-graphics-card-on-a-table-TErYPw4o1KM?utm_source=whatledto&utm_medium=referral",
    "description": "Nvidia RTX 2060",
    "license": "",
    "license_url": ""
  },
  "entities": [
    {
      "name": "AWS",
      "kind": "company",
      "role": "Cloud provider Nvidia named among the first expected deployers of Vera Rubin-based instances."
    },
    {
      "name": "Amazon Web Services",
      "kind": "company",
      "role": "Cloud provider NVIDIA named as an intended early deployer of Vera Rubin-based instances."
    },
    {
      "name": "CoreWeave",
      "kind": "company",
      "role": "Received its first production NVIDIA Vera Rubin NVL72 racks."
    },
    {
      "name": "Dell Technologies",
      "kind": "company",
      "role": "Delivered Vera Rubin NVL72 racks to CoreWeave using Dell infrastructure."
    },
    {
      "name": "Google Cloud",
      "kind": "company",
      "role": "Cloud provider NVIDIA named as an intended early deployer of Vera Rubin-based instances."
    },
    {
      "name": "Microsoft",
      "kind": "company",
      "role": "Customer reportedly told of a possible Blackwell delay and one of the cloud providers NVIDIA named for early Vera Rubin-based instances."
    },
    {
      "name": "Microsoft Azure",
      "kind": "company",
      "role": "Named by Nvidia as one of the first planned deployers of Vera Rubin-based instances."
    },
    {
      "name": "Nvidia",
      "kind": "company",
      "role": "Designer and seller of the Hopper, Blackwell and Rubin data center platforms and the source of the production and revenue disclosures."
    },
    {
      "name": "Oracle Cloud Infrastructure",
      "kind": "company",
      "role": "Cloud provider NVIDIA named as an intended early deployer of Vera Rubin-based instances."
    },
    {
      "name": "SemiAnalysis",
      "kind": "company",
      "role": "Analyst firm attributed by Tom’s Hardware with the reported Kyber NVL144 schedule claim."
    },
    {
      "name": "TSMC",
      "kind": "company",
      "role": "Chip manufacturer whose CoWoS packaging was cited by press reporting as a suspected, unconfirmed cause of the reported Blackwell shipment delay."
    },
    {
      "name": "Tom’s Hardware",
      "kind": "company",
      "role": "Published the articles reporting NVIDIA’s statements."
    },
    {
      "name": "Data Center",
      "kind": "other",
      "role": "NVIDIA business segment for which revenue was reported."
    },
    {
      "name": "The Information",
      "kind": "other",
      "role": "Outlet whose August 2024 report of a Blackwell delay triggered Nvidia's statements."
    },
    {
      "name": "Colette Kress",
      "kind": "person",
      "role": "NVIDIA chief financial officer who disclosed the Blackwell mask change intended to improve yields."
    },
    {
      "name": "Jensen Huang",
      "kind": "person",
      "role": "NVIDIA chief executive who presented the Rubin roadmap and said Vera Rubin was in full production."
    },
    {
      "name": "B200",
      "kind": "product",
      "role": "Blackwell Tensor Core GPU announced as a core component of the Blackwell platform."
    },
    {
      "name": "Blackwell",
      "kind": "product",
      "role": "Successor data center platform announced in March 2024, including the B200 GPU and GB200 Grace Blackwell Superchip."
    },
    {
      "name": "Blackwell Ultra",
      "kind": "product",
      "role": "Blackwell update announced for partner products in the second half of 2025; NVIDIA said its production was ramping in August 2025."
    },
    {
      "name": "Feynman",
      "kind": "product",
      "role": "Post-Rubin roadmap generation named in the event scope, with no timing or product details in the provided timeline nodes."
    },
    {
      "name": "GB200 Grace Blackwell Superchip",
      "kind": "product",
      "role": "A product pairing two B200 GPUs with a Grace CPU; it underpins the GB200 NVL72 system."
    },
    {
      "name": "GB300 NVL72",
      "kind": "product",
      "role": "Blackwell Ultra rack-scale system that Nvidia positioned as the successor to GB200 NVL72."
    },
    {
      "name": "Grace CPU",
      "kind": "product",
      "role": "Nvidia CPU used with Blackwell GPUs in the GB200 and followed by the Vera CPU in the Rubin platform."
    },
    {
      "name": "H100",
      "kind": "product",
      "role": "First Hopper GPU, announced in March 2022 and reported as shipping in November 2022."
    },
    {
      "name": "H200",
      "kind": "product",
      "role": "Hopper refresh using HBM3e memory, with systems expected to ship from the second quarter of 2024."
    },
    {
      "name": "Hopper",
      "kind": "product",
      "role": "NVIDIA data center architecture introduced in 2022, led by the H100 and later refreshed with the H200."
    },
    {
      "name": "Kyber NVL144",
      "kind": "product",
      "role": "Rack configuration discussed in the Tom’s Hardware article about the Rubin Ultra roadmap."
    },
    {
      "name": "NVL72",
      "kind": "product",
      "role": "Blackwell AI supercomputer that NVIDIA said was in full-scale production."
    },
    {
      "name": "Rubin",
      "kind": "product",
      "role": "2026 generation (Vera Rubin NVL144), with Rubin Ultra due in 2027."
    },
    {
      "name": "Rubin Ultra",
      "kind": "product",
      "role": "Follow-on Rubin system generation that NVIDIA scheduled for the second half of 2027."
    },
    {
      "name": "Vera",
      "kind": "product",
      "role": "Arm-based CPU paired with Rubin GPUs."
    },
    {
      "name": "Vera CPU",
      "kind": "product",
      "role": "Arm-based CPU announced with the Rubin platform and used in the Vera Rubin system."
    },
    {
      "name": "Vera Rubin",
      "kind": "product",
      "role": "NVIDIA's next-generation data center platform combining the Vera CPU, Rubin chips and associated networking and rack systems."
    },
    {
      "name": "Vera Rubin NVL72",
      "kind": "product",
      "role": "NVIDIA rack platform received by CoreWeave."
    }
  ],
  "nodes": [
    {
      "id": 1,
      "url": "https://whatledto.com/events/nvidia-datacenter-gpu-roadmap#n1",
      "date": "2022-03-22",
      "date_precision": "day",
      "title": "Nvidia announces the Hopper architecture and H100 GPU at GTC 2022",
      "summary": "Nvidia introduced Hopper, the successor to Ampere, with the H100 as its first GPU. The press release\ncites 80 billion transistors on a TSMC 4N process and says the H100 will be available in the third quarter.",
      "type": "announcement",
      "entities": [
        "Nvidia",
        "TSMC",
        "Hopper"
      ],
      "published_at": "2026-09-02T06:54:49.621274+00:00",
      "note": "",
      "image": "/images/nodes/nvidia-datacenter-gpu-roadmap/1-b2d80269.jpg",
      "image_credit": {
        "credit": "NVIDIA, press image",
        "source_url": "https://nvidianews.nvidia.com/news/nvidia-announces-hopper-architecture-the-next-generation-of-accelerated-computing",
        "image_url": "https://iprsoftwaremedia.com/219/files/20224/6296531eb3aed3741918b6b5_DH1L4415-HDR-20220527-r5/DH1L4415-HDR-20220527-r5_mid.jpg?v=dba29dde-5a9b-4ac8-bf58-a0b78da89547",
        "terms_url": "https://www.nvidia.com/en-us/about-nvidia/terms-of-service/",
        "width": 700,
        "height": 394,
        "mode": "editorial"
      },
      "sources": [
        {
          "url": "https://nvidianews.nvidia.com/news/nvidia-announces-hopper-architecture-the-next-generation-of-accelerated-computing",
          "outlet": "Nvidia Newsroom",
          "published": "2022-03-22",
          "quote": "NVIDIA H100 will be available starting in the third quarter."
        }
      ]
    },
    {
      "id": 14,
      "url": "https://whatledto.com/events/nvidia-datacenter-gpu-roadmap#n14",
      "date": "2022-11-16",
      "date_precision": "day",
      "title": "NVIDIA begins shipping Hopper-based H100 GPUs",
      "summary": "NVIDIA said it began shipping the Hopper-based H100 Tensor Core GPU. It said the first H100 systems were available.",
      "type": "shipment",
      "entities": [
        "Nvidia",
        "H100",
        "Hopper"
      ],
      "published_at": "2026-09-02T09:28:44.799200+00:00",
      "note": "The page is dated November 16, 2022, and its Data Center highlights state that NVIDIA began shipping the H100 Tensor Core GPU based on Hopper, with first systems available now.",
      "image": "",
      "image_credit": {
        "credit": "",
        "source_url": "",
        "image_url": "",
        "terms_url": "",
        "mode": "licensed"
      },
      "sources": [
        {
          "url": "https://nvidianews.nvidia.com/news/nvidia-announces-financial-results-for-third-quarter-fiscal-2023",
          "outlet": "NVIDIA Newsroom",
          "published": "2022-11-16",
          "quote": "Began shipping the NVIDIA® H100 Tensor Core GPU based on the new NVIDIA Hopper™ architecture, with first systems available now."
        }
      ]
    },
    {
      "id": 2,
      "url": "https://whatledto.com/events/nvidia-datacenter-gpu-roadmap#n2",
      "date": "2023-11-13",
      "date_precision": "day",
      "title": "Nvidia announces the H200, the first GPU with HBM3e memory",
      "summary": "The H200 kept the Hopper architecture but moved to HBM3e memory. Nvidia said systems from server makers\nand cloud providers would begin shipping in the second quarter of 2024.",
      "type": "announcement",
      "entities": [
        "Nvidia",
        "Hopper"
      ],
      "published_at": "2026-09-02T06:54:49.621276+00:00",
      "note": "",
      "image": "/images/nodes/nvidia-datacenter-gpu-roadmap/2-b2d80269.jpg",
      "image_credit": {
        "credit": "NVIDIA, press image",
        "source_url": "https://nvidianews.nvidia.com/news/nvidia-supercharges-hopper-the-worlds-leading-ai-computing-platform",
        "image_url": "https://iprsoftwaremedia.com/219/files/20224/6296531eb3aed3741918b6b5_DH1L4415-HDR-20220527-r5/DH1L4415-HDR-20220527-r5_mid.jpg?v=dba29dde-5a9b-4ac8-bf58-a0b78da89547",
        "terms_url": "https://www.nvidia.com/en-us/about-nvidia/terms-of-service/",
        "width": 700,
        "height": 394,
        "mode": "editorial"
      },
      "sources": [
        {
          "url": "https://nvidianews.nvidia.com/news/nvidia-supercharges-hopper-the-worlds-leading-ai-computing-platform",
          "outlet": "Nvidia Newsroom",
          "published": "2023-11-13",
          "quote": "With HBM3e, the NVIDIA H200 delivers 141GB of memory at 4.8 terabytes per second, nearly double the capacity and 2.4x more bandwidth compared with its predecessor."
        }
      ]
    },
    {
      "id": 3,
      "url": "https://whatledto.com/events/nvidia-datacenter-gpu-roadmap#n3",
      "date": "2024-03-18",
      "date_precision": "day",
      "title": "Blackwell platform announced at GTC 2024 with B200 and GB200",
      "summary": "Nvidia announced the Blackwell architecture in San Jose, listing the B200 Tensor Core GPU and the GB200\nGrace Blackwell Superchip, which pairs two B200 GPUs with a Grace CPU. The release claims up to 25x lower\ncost and energy for LLM inference compared with H100.",
      "type": "announcement",
      "entities": [
        "Nvidia",
        "Jensen Huang",
        "Blackwell",
        "TSMC"
      ],
      "published_at": "2026-09-02T06:54:49.621277+00:00",
      "note": "",
      "image": "",
      "image_credit": {
        "credit": "",
        "source_url": "",
        "image_url": "",
        "terms_url": "",
        "mode": "licensed"
      },
      "sources": [
        {
          "url": "https://investor.nvidia.com/news/press-release-details/2024/NVIDIA-Blackwell-Platform-Arrives-to-Power-a-New-Era-of-Computing/default.aspx",
          "outlet": "Nvidia Investor Relations",
          "published": "2024-03-18",
          "quote": "Packed with 208 billion transistors, Blackwell-architecture GPUs are manufactured"
        }
      ]
    },
    {
      "id": 16,
      "url": "https://whatledto.com/events/nvidia-datacenter-gpu-roadmap#n16",
      "date": "2024-05-22",
      "date_precision": "day",
      "title": "NVIDIA says Blackwell is in full production",
      "summary": "NVIDIA said the Blackwell platform was in full production. It said the platform formed the foundation for trillion-parameter-scale generative AI.",
      "type": "financial",
      "entities": [
        "Nvidia",
        "Blackwell"
      ],
      "published_at": "2026-09-02T09:28:44.799200+00:00",
      "note": "The page is readable and dated May 22, 2024. It states verbatim that the Blackwell platform was in full production and formed the foundation for trillion-parameter-scale generative AI.",
      "image": "",
      "image_credit": {
        "credit": "",
        "source_url": "",
        "image_url": "",
        "terms_url": "",
        "mode": "licensed"
      },
      "sources": [
        {
          "url": "https://investor.nvidia.com/news/press-release-details/2024/NVIDIA-Announces-Financial-Results-for-First-Quarter-Fiscal-2025/default.aspx",
          "outlet": "NVIDIA",
          "published": "2024-05-22",
          "quote": "We are poised for our next wave of growth. The Blackwell platform is in full production and forms the foundation for trillion-parameter-scale generative AI."
        }
      ]
    },
    {
      "id": 4,
      "url": "https://whatledto.com/events/nvidia-datacenter-gpu-roadmap#n4",
      "date": "2024-06-02",
      "date_precision": "day",
      "title": "Rubin platform and Vera CPU revealed ahead of Computex 2024",
      "summary": "In his Computex keynote in Taipei, Jensen Huang showed the Rubin platform as Blackwell's successor,\ntogether with an Arm-based CPU called Vera and new networking parts. He described a one-year release\nrhythm for the whole data center stack. No launch year was given in Nvidia's own write-up.",
      "type": "announcement",
      "entities": [
        "Nvidia",
        "Jensen Huang",
        "Rubin",
        "Vera"
      ],
      "published_at": "2026-09-02T06:54:49.621277+00:00",
      "note": "",
      "image": "",
      "image_credit": {
        "credit": "",
        "source_url": "",
        "image_url": "",
        "terms_url": "",
        "mode": "licensed"
      },
      "sources": [
        {
          "url": "https://blogs.nvidia.com/blog/computex-2024-jensen-huang/",
          "outlet": "Nvidia Blog",
          "published": "2024-06-02",
          "quote": "Revealed for the first time, the Rubin platform will succeed the upcoming Blackwell platform, featuring new GPUs, a new Arm-based CPU — Vera — and advanced networking with NVLink 6, CX9 SuperNIC and the X1600 converged InfiniBand/Ethernet switch."
        }
      ]
    },
    {
      "id": 5,
      "url": "https://whatledto.com/events/nvidia-datacenter-gpu-roadmap#n5",
      "date": "2024-08-05",
      "date_precision": "day",
      "title": "Blackwell shipments reported delayed to early 2025; Nvidia says production is on track",
      "summary": "Citing The Information, The Register reported that Nvidia had told Microsoft of delays to the most\nadvanced Blackwell parts, with volume shipments pushed to the first quarter of 2025 and TSMC's CoWoS\npackaging suspected as the cause. Nvidia's statement did not confirm a delay: it said sampling had started\nand production was on track to ramp in the second half.",
      "type": "report",
      "entities": [
        "Nvidia",
        "Microsoft",
        "TSMC",
        "The Information",
        "Blackwell"
      ],
      "published_at": "2026-09-02T06:54:49.621278+00:00",
      "note": "",
      "image": "",
      "image_credit": {
        "credit": "",
        "source_url": "",
        "image_url": "",
        "terms_url": "",
        "mode": "licensed"
      },
      "sources": [
        {
          "url": "https://www.theregister.com/2024/08/05/nvidia_delays_blackwell_gpus_until/",
          "outlet": "The Register",
          "published": "2024-08-05",
          "quote": "As we've stated before, Hopper demand is very strong, broad Blackwell sampling has started, and production is on track to ramp in 2H. Beyond that, we don't comment on rumors."
        }
      ]
    },
    {
      "id": 6,
      "url": "https://whatledto.com/events/nvidia-datacenter-gpu-roadmap#n6",
      "date": "2024-08-28",
      "date_precision": "day",
      "title": "Nvidia discloses a Blackwell mask change to improve yields on its Q2 FY2025 call",
      "summary": "On the earnings call, CFO Colette Kress said Nvidia had changed the Blackwell GPU mask to improve\nproduction yields and that the production ramp would begin in the fourth quarter. Jensen Huang said\nNvidia meant to ship product, not merely start production, in Q4. Nvidia also said Blackwell demand was\nwell above supply.",
      "type": "response",
      "entities": [
        "Nvidia",
        "Colette Kress",
        "Jensen Huang",
        "Blackwell"
      ],
      "published_at": "2026-09-02T06:54:49.621278+00:00",
      "note": "",
      "image": "",
      "image_credit": {
        "credit": "",
        "source_url": "",
        "image_url": "",
        "terms_url": "",
        "mode": "licensed"
      },
      "sources": [
        {
          "url": "https://www.theregister.com/2024/08/29/nvidia_blackwell_manufacturing/",
          "outlet": "The Register",
          "published": "2024-08-29",
          "quote": "We executed a change to the Blackwell GPU mask to improve production yields."
        }
      ]
    },
    {
      "id": 17,
      "url": "https://whatledto.com/events/nvidia-datacenter-gpu-roadmap#n17",
      "date": "2024-08-28",
      "date_precision": "day",
      "title": "NVIDIA says Blackwell samples are shipping",
      "summary": "NVIDIA said Blackwell samples were shipping to partners and customers. The company identified the products as Blackwell samples.",
      "type": "shipment",
      "entities": [
        "Nvidia",
        "Blackwell"
      ],
      "published_at": "2026-09-02T09:28:44.799200+00:00",
      "note": "The page is readable and dated August 28, 2024. It states, “Blackwell samples are shipping to our partners and customers,” supporting the summary as written.",
      "image": "",
      "image_credit": {
        "credit": "",
        "source_url": "",
        "image_url": "",
        "terms_url": "",
        "mode": "licensed"
      },
      "sources": [
        {
          "url": "https://investor.nvidia.com/news/press-release-details/2024/NVIDIA-Announces-Financial-Results-for-Second-Quarter-Fiscal-2025/default.aspx",
          "outlet": "NVIDIA",
          "published": "2024-08-28",
          "quote": "Blackwell samples are shipping to our partners and customers."
        }
      ]
    },
    {
      "id": 7,
      "url": "https://whatledto.com/events/nvidia-datacenter-gpu-roadmap#n7",
      "date": "2024-11-20",
      "date_precision": "day",
      "title": "Nvidia says Blackwell is in full production; data center revenue hits $30.8 billion",
      "summary": "In its third-quarter fiscal 2025 results, Nvidia described Blackwell as being in full production and\nreported record data center revenue of $30.8 billion, up 17% from the prior quarter.",
      "type": "financial",
      "entities": [
        "Nvidia",
        "Jensen Huang",
        "Blackwell",
        "Hopper"
      ],
      "published_at": "2026-09-02T06:54:49.621279+00:00",
      "note": "",
      "image": "",
      "image_credit": {
        "credit": "",
        "source_url": "",
        "image_url": "",
        "terms_url": "",
        "mode": "licensed"
      },
      "sources": [
        {
          "url": "https://nvidianews.nvidia.com/news/nvidia-announces-financial-results-for-third-quarter-fiscal-2025",
          "outlet": "Nvidia Newsroom",
          "published": "2024-11-20",
          "quote": "Third-quarter revenue was a record $30.8 billion, up 17% from the previous quarter"
        }
      ]
    },
    {
      "id": 8,
      "url": "https://whatledto.com/events/nvidia-datacenter-gpu-roadmap#n8",
      "date": "2025-02-26",
      "date_precision": "day",
      "title": "First quarter of Blackwell sales: Nvidia cites billions in revenue, data center at $35.6 billion",
      "summary": "Fourth-quarter fiscal 2025 results reported data center revenue of $35.6 billion. Nvidia said Blackwell\nAI supercomputers achieved billions of dollars in sales in their first quarter, without giving an exact\nfigure in the release.",
      "type": "financial",
      "entities": [
        "Nvidia",
        "Jensen Huang",
        "Blackwell"
      ],
      "published_at": "2026-09-02T06:54:49.621279+00:00",
      "note": "",
      "image": "",
      "image_credit": {
        "credit": "",
        "source_url": "",
        "image_url": "",
        "terms_url": "",
        "mode": "licensed"
      },
      "sources": [
        {
          "url": "https://nvidianews.nvidia.com/news/nvidia-announces-financial-results-for-fourth-quarter-and-fiscal-2025",
          "outlet": "Nvidia Newsroom",
          "published": "2025-02-26",
          "quote": "Fourth-quarter revenue was a record $35.6 billion, up 16% from the previous quarter and up 93% from a year ago."
        }
      ]
    },
    {
      "id": 9,
      "url": "https://whatledto.com/events/nvidia-datacenter-gpu-roadmap#n9",
      "date": "2025-03-18",
      "date_precision": "day",
      "title": "Blackwell Ultra announced at GTC 2025 for the second half of 2025",
      "summary": "Nvidia announced Blackwell Ultra and the GB300 NVL72 rack, claiming 1.5x the AI performance of the GB200\nNVL72. Partner products were expected from the second half of 2025.",
      "type": "announcement",
      "entities": [
        "Nvidia",
        "Blackwell"
      ],
      "published_at": "2026-09-02T06:54:49.621280+00:00",
      "note": "",
      "image": "",
      "image_credit": {
        "credit": "",
        "source_url": "",
        "image_url": "",
        "terms_url": "",
        "mode": "licensed"
      },
      "sources": [
        {
          "url": "https://investor.nvidia.com/news/press-release-details/2025/NVIDIA-Blackwell-Ultra-AI-Factory-Platform-Paves-Way-for-Age-of-AI-Reasoning/default.aspx",
          "outlet": "Nvidia Investor Relations",
          "published": "2025-03-18",
          "quote": "Blackwell Ultra-based products are expected to be available from partners starting from the second half of 2025."
        }
      ]
    },
    {
      "id": 10,
      "url": "https://whatledto.com/events/nvidia-datacenter-gpu-roadmap#n10",
      "date": "2025-03-18",
      "date_precision": "day",
      "title": "Roadmap set: Vera Rubin NVL144 in H2 2026, Rubin Ultra in H2 2027",
      "summary": "In the GTC 2025 keynote, Nvidia dated the next two generations. Vera Rubin NVL144 systems were slated\nfor the second half of 2026 and Rubin Ultra systems for the second half of 2027.",
      "type": "announcement",
      "entities": [
        "Nvidia",
        "Jensen Huang",
        "Rubin",
        "Vera"
      ],
      "published_at": "2026-09-02T06:54:49.621280+00:00",
      "note": "",
      "image": "",
      "image_credit": {
        "credit": "",
        "source_url": "",
        "image_url": "",
        "terms_url": "",
        "mode": "licensed"
      },
      "sources": [
        {
          "url": "https://blogs.nvidia.com/blog/nvidia-keynote-at-gtc-2025-ai-news-live-updates/",
          "outlet": "Nvidia Blog",
          "published": "2025-03-18",
          "quote": "Systems built on Rubin Ultra, including the Vera Rubin NVL 144, will arrive in the second half of next year. And due for the second half of 2027: systems built on Rubin Ultra."
        }
      ]
    },
    {
      "id": 18,
      "url": "https://whatledto.com/events/nvidia-datacenter-gpu-roadmap#n18",
      "date": "2025-05-28",
      "date_precision": "day",
      "title": "NVIDIA says Blackwell NVL72 is in full-scale production",
      "summary": "NVIDIA said its Blackwell NVL72 AI supercomputer was in full-scale production. It said production covered system makers and cloud service providers.",
      "type": "shipment",
      "entities": [
        "Nvidia",
        "Blackwell",
        "NVL72"
      ],
      "published_at": "2026-09-02T09:28:44.799200+00:00",
      "note": "The page, dated May 28, 2025, states that NVIDIA's Blackwell NVL72 AI supercomputer is in full-scale production across system makers and cloud service providers. The quoted statement appears on the page.",
      "image": "/images/nodes/nvidia-datacenter-gpu-roadmap/18-b7612702.jpg",
      "image_credit": {
        "credit": "NVIDIA, press image",
        "source_url": "https://nvidianews.nvidia.com/news/nvidia-announces-financial-results-for-first-quarter-fiscal-2026",
        "image_url": "https://iprsoftwaremedia.com/219/files/202502/67bf85d33d633281c6875f53_aabe86db-ce89-4434-b83c-49508297980/aabe86db-ce89-4434-b83c-49508297980_615f1395-459d-4e22-a907-06dcd0750e3b-prv.jpg?v=615f1395-459d-4e22-a907-06dcd0750e3b",
        "terms_url": "https://www.nvidia.com/en-us/about-nvidia/terms-of-service/",
        "width": 1204,
        "height": 801,
        "mode": "editorial"
      },
      "sources": [
        {
          "url": "https://nvidianews.nvidia.com/news/nvidia-announces-financial-results-for-first-quarter-fiscal-2026",
          "outlet": "NVIDIA",
          "published": "2025-05-28",
          "quote": "Our breakthrough Blackwell NVL72 AI supercomputer — a ‘thinking machine’ designed for reasoning— is now in full-scale production across system makers and cloud service providers."
        }
      ]
    },
    {
      "id": 19,
      "url": "https://whatledto.com/events/nvidia-datacenter-gpu-roadmap#n19",
      "date": "2025-08-27",
      "date_precision": "day",
      "title": "NVIDIA says Blackwell Ultra production is ramping",
      "summary": "NVIDIA said production of Blackwell Ultra was ramping at full speed. It identified Blackwell as its AI platform.",
      "type": "financial",
      "entities": [
        "Nvidia",
        "Blackwell",
        "Blackwell Ultra"
      ],
      "published_at": "2026-09-02T09:28:44.799200+00:00",
      "note": "The page is readable and is dated August 27, 2025. It contains the quoted statement verbatim and identifies Blackwell as the AI platform while stating that Blackwell Ultra production is ramping at full speed.",
      "image": "",
      "image_credit": {
        "credit": "",
        "source_url": "",
        "image_url": "",
        "terms_url": "",
        "mode": "licensed"
      },
      "sources": [
        {
          "url": "https://investor.nvidia.com/news/press-release-details/2025/NVIDIA-Announces-Financial-Results-for-Second-Quarter-Fiscal-2026/default.aspx",
          "outlet": "NVIDIA",
          "published": "2025-08-27",
          "quote": "Blackwell is the AI platform the world has been waiting for, delivering an exceptional generational leap — production of Blackwell Ultra is ramping at full speed, and demand is extraordinary."
        }
      ]
    },
    {
      "id": 20,
      "url": "https://whatledto.com/events/nvidia-datacenter-gpu-roadmap#n20",
      "date": "2025-11-19",
      "date_precision": "day",
      "title": "NVIDIA says Blackwell sales and cloud GPUs are sold out",
      "summary": "NVIDIA reported Blackwell sales. It said cloud GPUs were sold out.",
      "type": "financial",
      "entities": [
        "Nvidia",
        "Blackwell"
      ],
      "published_at": "2026-09-02T09:28:44.799200+00:00",
      "note": "The page is readable and dated November 19, 2025. It reports Blackwell sales and states that cloud GPUs are sold out, using the quoted wording.",
      "image": "",
      "image_credit": {
        "credit": "",
        "source_url": "",
        "image_url": "",
        "terms_url": "",
        "mode": "licensed"
      },
      "sources": [
        {
          "url": "https://investor.nvidia.com/news/press-release-details/2025/NVIDIA-Announces-Financial-Results-for-Third-Quarter-Fiscal-2026/default.aspx",
          "outlet": "NVIDIA",
          "published": "2025-11-19",
          "quote": "Blackwell sales are off the charts, and cloud GPUs are sold out."
        }
      ]
    },
    {
      "id": 21,
      "url": "https://whatledto.com/events/nvidia-datacenter-gpu-roadmap#n21",
      "date": "2026-02-25",
      "date_precision": "day",
      "title": "NVIDIA unveils the six-chip Rubin platform",
      "summary": "NVIDIA unveiled the Rubin platform, which it said comprised six chips. It said AWS, Google Cloud, Microsoft Azure and Oracle Cloud Infrastructure would be among the first to deploy Vera Rubin-based instances.",
      "type": "financial",
      "entities": [
        "Nvidia",
        "Rubin",
        "Vera Rubin",
        "Blackwell",
        "AWS",
        "Google Cloud",
        "Microsoft Azure",
        "Oracle Cloud Infrastructure"
      ],
      "published_at": "2026-09-02T09:28:44.799200+00:00",
      "note": "The page is readable and dated February 25, 2026. Its Data Center highlights state that NVIDIA unveiled the Rubin platform comprising six new chips and name AWS, Google Cloud, Microsoft Azure, and Oracle Cloud Infrastructure as early deployers of Vera Rubin-based instances.",
      "image": "",
      "image_credit": {
        "credit": "",
        "source_url": "",
        "image_url": "",
        "terms_url": "",
        "mode": "licensed"
      },
      "sources": [
        {
          "url": "https://investor.nvidia.com/news/press-release-details/2026/NVIDIA-Announces-Financial-Results-for-Fourth-Quarter-and-Fiscal-2026/",
          "outlet": "NVIDIA",
          "published": "2026-02-25",
          "quote": "Unveiled the NVIDIA Rubin platform, comprising six new chips to deliver up to a 10x reduction in inference token cost, compared with the NVIDIA Blackwell platform; cloud providers Amazon Web Services (AWS), Google Cloud, Microsoft Azure and Oracle Cloud Infrastructure will be among the first to deploy Vera Rubin-based instances."
        }
      ]
    },
    {
      "id": 11,
      "url": "https://whatledto.com/events/nvidia-datacenter-gpu-roadmap#n11",
      "date": "2026-03-16",
      "date_precision": "day",
      "title": "Vera Rubin declared in full production at GTC 2026",
      "summary": "Nvidia announced that the seven chips of the Vera Rubin platform were in full production and described\nthe platform as five racks and one supercomputer. This kept to the second-half-2026 timing set a year\nearlier.",
      "type": "announcement",
      "entities": [
        "Nvidia",
        "Jensen Huang",
        "Rubin",
        "Vera"
      ],
      "published_at": "2026-09-02T06:54:49.621281+00:00",
      "note": "",
      "image": "/images/nodes/nvidia-datacenter-gpu-roadmap/11-5b5e1c54.jpg",
      "image_credit": {
        "credit": "NVIDIA, press image",
        "source_url": "https://nvidianews.nvidia.com/news/nvidia-vera-rubin-platform",
        "image_url": "https://iprsoftwaremedia.com/219/files/202608/01129f086f93bf1c2014fee1bc8cd4e2/6a8c7ec03d6332d2f389b07b_jetson-orin-nano-2/jetson-orin-nano-2_mid.jpg?v=14abfacb-ea35-4a1f-a3d4-01693db66823",
        "terms_url": "https://www.nvidia.com/en-us/about-nvidia/terms-of-service/",
        "width": 700,
        "height": 393,
        "mode": "editorial"
      },
      "sources": [
        {
          "url": "https://nvidianews.nvidia.com/news/nvidia-vera-rubin-platform",
          "outlet": "Nvidia Newsroom",
          "published": "2026-03-16",
          "quote": "Vera Rubin is a generational leap — seven breakthrough chips, five racks, one giant supercomputer — built to power every phase of AI"
        }
      ]
    },
    {
      "id": 12,
      "url": "https://whatledto.com/events/nvidia-datacenter-gpu-roadmap#n12",
      "date": "2026-05-31",
      "date_precision": "day",
      "title": "Vera Rubin ramping across the supply chain; production shipments set for fall 2026",
      "summary": "At GTC Taipei ahead of Computex, Nvidia said hundreds of supply chain partners across more than 350\nfactories in 30 countries were ramping Vera Rubin, with production shipments to begin in the fall.",
      "type": "announcement",
      "entities": [
        "Nvidia",
        "Rubin",
        "TSMC"
      ],
      "published_at": "2026-09-02T06:54:49.621281+00:00",
      "note": "",
      "image": "/images/nodes/nvidia-datacenter-gpu-roadmap/12-5b5e1c54.jpg",
      "image_credit": {
        "credit": "NVIDIA, press image",
        "source_url": "https://nvidianews.nvidia.com/news/vera-rubin-full-production-agentic-ai-factory",
        "image_url": "https://iprsoftwaremedia.com/219/files/202608/01129f086f93bf1c2014fee1bc8cd4e2/6a8c7ec03d6332d2f389b07b_jetson-orin-nano-2/jetson-orin-nano-2_mid.jpg?v=14abfacb-ea35-4a1f-a3d4-01693db66823",
        "terms_url": "https://www.nvidia.com/en-us/about-nvidia/terms-of-service/",
        "width": 700,
        "height": 393,
        "mode": "editorial"
      },
      "sources": [
        {
          "url": "https://nvidianews.nvidia.com/news/vera-rubin-full-production-agentic-ai-factory",
          "outlet": "Nvidia Newsroom",
          "published": "2026-05-31",
          "quote": "Production shipments of Vera Rubin are set to begin starting this fall."
        }
      ]
    },
    {
      "id": 663,
      "url": "https://whatledto.com/events/nvidia-datacenter-gpu-roadmap#n663",
      "date": "2026-07-06",
      "date_precision": "day",
      "title": "NVIDIA says its roadmap is intact",
      "summary": "NVIDIA said, “Our roadmap is intact.” Tom’s Hardware published the statement in an article about the Rubin Ultra Kyber NVL144 rack.",
      "type": "response",
      "entities": [
        "Nvidia",
        "Rubin Ultra",
        "Kyber NVL144",
        "SemiAnalysis",
        "Tom’s Hardware"
      ],
      "published_at": "2026-09-09T01:01:52.511727+00:00",
      "note": "The article’s update dated July 6, 2026 says an Nvidia representative told Tom’s Hardware, “Our roadmap is intact.” It appears in an article about the reported delay of the Kyber NVL144 rack for Rubin Ultra.",
      "image": "",
      "image_credit": {
        "credit": "",
        "source_url": "",
        "image_url": "",
        "terms_url": "",
        "mode": "licensed"
      },
      "sources": [
        {
          "url": "https://www.tomshardware.com/pc-components/gpus/nvidias-kyber-rack-for-rubin-ultra-slips-to-2028",
          "outlet": "Tom’s Hardware",
          "published": "2026-07-06",
          "quote": "“Our roadmap is intact.”"
        }
      ]
    },
    {
      "id": 13,
      "url": "https://whatledto.com/events/nvidia-datacenter-gpu-roadmap#n13",
      "date": "2026-08-26",
      "date_precision": "day",
      "title": "Nvidia reports $96.2 billion quarter, data center $89.0 billion, Vera Rubin in full production",
      "summary": "Second-quarter fiscal 2027 results showed revenue of $96.2 billion, up 106% year on year, with data\ncenter revenue of $89.0 billion. Jensen Huang said Vera Rubin was in full production, and the release\nnoted racks running at partners.",
      "type": "financial",
      "entities": [
        "Nvidia",
        "Jensen Huang",
        "Rubin",
        "Blackwell"
      ],
      "published_at": "2026-09-02T06:54:49.621281+00:00",
      "note": "",
      "image": "",
      "image_credit": {
        "credit": "",
        "source_url": "",
        "image_url": "",
        "terms_url": "",
        "mode": "licensed"
      },
      "sources": [
        {
          "url": "https://nvidianews.nvidia.com/news/nvidia-announces-financial-results-for-second-quarter-fiscal-2027",
          "outlet": "Nvidia Newsroom",
          "published": "2026-08-26",
          "quote": "Vera Rubin, now in full production, was built to power exactly this moment."
        }
      ]
    },
    {
      "id": 23,
      "url": "https://whatledto.com/events/nvidia-datacenter-gpu-roadmap#n23",
      "date": "2026-08-27",
      "date_precision": "day",
      "title": "NVIDIA says Vera Rubin began production shipments in fiscal Q3 2027",
      "summary": "NVIDIA stated that Vera Rubin began production shipments in the third quarter of fiscal 2027. The filing described Vera Rubin as its next-generation Data Center architecture.",
      "type": "regulatory",
      "entities": [
        "Nvidia",
        "Vera Rubin"
      ],
      "published_at": "2026-09-02T09:28:44.799200+00:00",
      "note": "The filing states that Vera Rubin, NVIDIA’s next-generation Data Center architecture, began production shipments in the third quarter of fiscal year 2027. The filing supports the node date and does not give a different event date.",
      "image": "",
      "image_credit": {
        "credit": "",
        "source_url": "",
        "image_url": "",
        "terms_url": "",
        "mode": "licensed"
      },
      "sources": [
        {
          "url": "https://investor.nvidia.com/files/doc_financials/2027/NVDA-2027-Q2-10Q-Final-including-exhibits.pdf",
          "outlet": "NVIDIA Form 10-Q",
          "published": "2026-08-27",
          "quote": "Our next-generation Data Center architecture, Vera Rubin, began production shipments in the third quarter of fiscal year 2027."
        }
      ]
    }
  ],
  "relations": [
    {
      "from": 5,
      "to": 6,
      "kind": "responds_to",
      "note": "Three weeks after saying production was on track, Nvidia acknowledged a mask change to fix Blackwell yields."
    },
    {
      "from": 6,
      "to": 7,
      "kind": "follows",
      "note": "The Q4 ramp promised in August was described as full production by November."
    },
    {
      "from": 7,
      "to": 8,
      "kind": "follows",
      "note": "Full production in November led to the first quarter of Blackwell revenue, reported in February."
    },
    {
      "from": 4,
      "to": 10,
      "kind": "follows",
      "note": "Rubin was shown without a date at Computex 2024; GTC 2025 attached second-half-2026 timing."
    },
    {
      "from": 10,
      "to": 11,
      "kind": "follows",
      "note": "Full production announced in March 2026 is consistent with the second-half-2026 schedule."
    },
    {
      "from": 11,
      "to": 12,
      "kind": "follows",
      "note": "Chips in production in March; racks ramping at partners and shipments dated for fall by May."
    },
    {
      "from": 12,
      "to": 13,
      "kind": "follows",
      "note": "August results repeated the full-production status and cited racks running at partners."
    },
    {
      "from": 16,
      "to": 17,
      "kind": "follows",
      "note": "NVIDIA first said Blackwell was in full production and later said Blackwell samples were shipping to partners and customers."
    },
    {
      "from": 17,
      "to": 18,
      "kind": "follows",
      "note": "After reporting Blackwell sample shipments, NVIDIA later said Blackwell NVL72 was in full-scale production across system makers and cloud service providers."
    },
    {
      "from": 18,
      "to": 19,
      "kind": "follows",
      "note": "After NVIDIA described Blackwell NVL72 as in full-scale production, it said Blackwell Ultra production was ramping at full speed."
    },
    {
      "from": 19,
      "to": 20,
      "kind": "follows",
      "note": "After reporting that Blackwell Ultra production was ramping, NVIDIA later said Blackwell sales were occurring and cloud GPUs were sold out."
    },
    {
      "from": 20,
      "to": 21,
      "kind": "follows",
      "note": "After its November 2025 Blackwell sales statement, NVIDIA's February 2026 release unveiled the Rubin platform."
    },
    {
      "from": 12,
      "to": 23,
      "kind": "contradicts",
      "note": "The existing node said Vera Rubin production shipments were set for fall 2026, while NVIDIA's August 2026 filing said shipments had begun in fiscal Q3 2027."
    }
  ],
  "links": [],
  "key_node_ids": [
    1,
    14,
    3,
    6,
    8,
    9,
    10,
    11,
    12,
    13,
    23
  ],
  "catalysts": [
    {
      "date": "2027-03-15",
      "title": "GTC 2027 in San Jose",
      "why": "NVIDIA's annual conference is a scheduled venue where it could provide further details on Rubin Ultra and the Feynman roadmap.",
      "source_url": "https://www.nvidia.com/gtc/"
    },
    {
      "date": "2027-07",
      "title": "Rubin Ultra enters NVIDIA's stated second-half-2027 window",
      "why": "NVIDIA has scheduled Rubin Ultra systems for the second half of 2027, making this the next disclosed platform transition after Vera Rubin.",
      "source_url": "https://blogs.nvidia.com/blog/nvidia-keynote-at-gtc-2025-ai-news-live-updates/"
    }
  ],
  "open_questions": [
    "NVIDIA said Blackwell was in full production in May 2024, but later described samples shipping in August and disclosed a mask change to improve yields; the disclosures do not make clear which products or production stages each statement covered.",
    "NVIDIA said in May 2026 that Vera Rubin production shipments would begin in the fall, while its August filing said shipments began in fiscal Q3 2027; the sources do not provide a precise calendar shipment date that resolves the apparent discrepancy.",
    "NVIDIA described Rubin as a six-chip platform in February 2026 and as a seven-chip Vera Rubin platform in March 2026, without explaining the difference in those counts.",
    "NVIDIA said its roadmap was intact after reporting on a claim that the Rubin Ultra Kyber NVL144 rack had slipped to 2028, but it did not specify whether Rubin Ultra systems remain on the previously stated second-half-2027 schedule.",
    "The provided sources do not state Feynman's specifications, production status, or launch timing."
  ],
  "metrics": [
    {
      "key": "dc-revenue",
      "name": "Nvidia data center revenue in the quarter",
      "unit": "$B",
      "kind": "bar",
      "points": [
        {
          "node_id": 7,
          "date": "2024-11-20",
          "value": 30.8,
          "quote": "reported record data center revenue of $30.8 billion, up 17% from the prior quarter."
        },
        {
          "node_id": 8,
          "date": "2025-02-26",
          "value": 35.6,
          "quote": "Fourth-quarter fiscal 2025 results reported data center revenue of $35.6 billion."
        },
        {
          "node_id": 13,
          "date": "2026-08-26",
          "value": 89,
          "quote": "with data center revenue of $89.0 billion."
        }
      ]
    }
  ],
  "related": [
    {
      "slug": "hbm-memory-supply",
      "url": "https://whatledto.com/events/hbm-memory-supply",
      "json": "https://whatledto.com/events/hbm-memory-supply.json"
    },
    {
      "slug": "nvidia-china-export-controls",
      "url": "https://whatledto.com/events/nvidia-china-export-controls",
      "json": "https://whatledto.com/events/nvidia-china-export-controls.json"
    },
    {
      "slug": "amd-instinct-ai-gpus",
      "url": "https://whatledto.com/events/amd-instinct-ai-gpus",
      "json": "https://whatledto.com/events/amd-instinct-ai-gpus.json"
    }
  ],
  "changelog": [
    {
      "date": "2026-09-09T01:02:40.072271+00:00",
      "kind": "summary",
      "summary": "Chart data refreshed: 3 point(s) across 1 metric(s).",
      "node_ids": []
    },
    {
      "date": "2026-09-09T01:02:40.072265+00:00",
      "kind": "summary",
      "summary": "Summary, key players and catalysts refreshed.",
      "node_ids": []
    },
    {
      "date": "2026-09-09T01:01:52.557803+00:00",
      "kind": "publish",
      "summary": "Added 1 node(s): 2026-07-06 NVIDIA says its roadmap is intact.",
      "node_ids": [
        663
      ]
    },
    {
      "date": "2026-09-09T00:54:06.309887+00:00",
      "kind": "research",
      "summary": "Research run found 3 candidate node(s).",
      "node_ids": [
        663,
        664,
        665
      ]
    },
    {
      "date": "2026-09-07T06:15:43.464547+00:00",
      "kind": "summary",
      "summary": "Chart data refreshed: 3 point(s) across 1 metric(s).",
      "node_ids": []
    },
    {
      "date": "2026-09-07T05:29:50.152867+00:00",
      "kind": "summary",
      "summary": "Summary, key players and catalysts refreshed.",
      "node_ids": []
    },
    {
      "date": "2026-09-02T09:33:02.753340+00:00",
      "kind": "summary",
      "summary": "Summary, key players and catalysts refreshed.",
      "node_ids": []
    },
    {
      "date": "2026-09-02T09:29:15.563125+00:00",
      "kind": "summary",
      "summary": "Summary, key players and catalysts refreshed.",
      "node_ids": []
    },
    {
      "date": "2026-09-02T09:28:44.806824+00:00",
      "kind": "publish",
      "summary": "Added 8 node(s): 2022-11-16 NVIDIA begins shipping Hopper-based H100 GPUs; 2024-05-22 NVIDIA says Blackwell is in full production; 2024-08-28 NVIDIA says Blackwell samples are shipping; 2025-05-28 NVIDIA says Blackwell NVL72 is in full-scale production and 4 more.",
      "node_ids": [
        14,
        16,
        17,
        18,
        19,
        20,
        21,
        23
      ]
    },
    {
      "date": "2026-09-02T09:02:55.471553+00:00",
      "kind": "research",
      "summary": "Research run found 8 candidate node(s).",
      "node_ids": [
        16,
        17,
        18,
        19,
        20,
        21,
        22,
        23
      ]
    },
    {
      "date": "2026-09-02T08:57:41.072325+00:00",
      "kind": "research",
      "summary": "Research run found 2 candidate node(s).",
      "node_ids": [
        14,
        15
      ]
    },
    {
      "date": "2026-09-02T06:54:49.628225+00:00",
      "kind": "seed",
      "summary": "Seeded 13 hand-verified node(s) from nvidia-datacenter-gpu-roadmap.yaml.",
      "node_ids": [
        1,
        2,
        3,
        4,
        5,
        6,
        7,
        8,
        9,
        10,
        11,
        12,
        13
      ]
    }
  ],
  "license": {
    "holder": "What Led To",
    "terms": "https://whatledto.com/methodology#reuse",
    "summary": "Quote entries with attribution and a link to the entry; do not republish whole timelines or train models on the text."
  }
}