Nvidia's Next-Gen GPUs are rumored to be codenamed 'Rubin,' arrive in 2025

Nvidia Hopper H100 die shot
(Image credit: Nvidia)

The codename of Nvidia's post-Blackwell GPU architecture could be Vera Rubin, if a new statement by hardware leaker @kopite7kimi is correct. Given Nvidia's new and more aggressive introduction cadence, Nvidia's Rubin GPUs for artificial intelligence (AI) and high-performance computing (HPC) are expected to be released in 2025. However, it is unclear whether Nvidia's client GPUs are set to adopt the same architecture.

@kopite7kimi says two GPUs belong to the Rubin family, the R100 and the GR200. The R100 is likely the first product based on the 'big' Rubin GPU for AI and HPC workloads, whereas the GR200 is likely a refined Rubin GPU akin to the GH200 GPU launched earlier this month. It is close to impossible to guess the characteristics of these GPUs, given that they are two years away and their target specifications are only being finalized now.

Related GPU Deals

TOC

(Image credit: Tom's Hardware)

Nvidia RTX 4060 Ti: now $449
RTX 4090 Prices: from $2249
Radeon RX 7800 XT: now $519
Radeon RX 6600: now $189

In fact, we would not even try to guess whether Nvidia plans to make Rubin GPUs at its traditional foundry partner TSMC (using its N3P or N3X fabrication process), Intel (using its 18A manufacturing technology), or Samsung (using its SF3 or SF3P production node).

Nvidia's newly unveiled roadmap indicates that from now on, the company will update its data center-grade GPUs aimed at AI and HPC workloads once a year, not once in two years as previously. As a result, the company aims to release codenamed Blackwell products in 2024, and codenamed Rubin products in 2025. It is unclear whether the company also plans to update its client GPUs once a year, like in the good old days of the 2000s and 2010s. To that end, while we expect Blackwell to power some of the best graphics cards in 2024 and 2025, we suspect the Vera Rubin architecture may not make it to the consumer market — it would be more like the Volta architecture.

It is noteworthy that Vera Rubin will be the first of Nvidia's architectures named after an astronomer. Through her observations of the rotation rates of galaxies in the 1960s and 1970s, Rubin provided strong evidence for the existence of dark matter, which was a major contribution to the field of physics. Previously Nvidia named its GPU architectures only after physicists, including Fahrenheit, Celsius, Kelvin, Rankine, Curie, Tesla, Fermi, Kepler, Maxwell, Pascal, Volta, Turing, Ampere, Ada Lovelace, Hopper, and Blackwell.

TOPICS
Anton Shilov
Contributing Writer

Anton Shilov is a contributing writer at Tom’s Hardware. Over the past couple of decades, he has covered everything from CPUs and GPUs to supercomputers and from modern process technologies and latest fab tools to high-tech industry trends.

  • phatboe
    Alan Turing was not a physicist either.
    Reply
  • TJ Hooker
    Admin said:
    It is noteworthy that Vera Rubin will be the first of Nvidia's architectures named after an astronomer.
    Previously Nvidia named its GPU architectures only after physicists, including Fahrenheit, Celsius, Kelvin, Rankine, Curie, Tesla, Fermi, Kepler, Maxwell, Pascal, Volta, Turing, Ampere, Ada Lovelace, Hopper, and Blackwell.
    Celsius and Keplar were also astronomers (although, in general, the distinction between astronomer and physicist can be a bit blurry). And several others weren't physicists (or weren't primarily physicists), e.g. 4 of the last 5.
    Reply
  • valthuer
    @JarredWaltonGPU Sorry to bother you about this, but, in light of these news, when exactly will the RTX-50 series be released?

    Does the initial Nvidia roadmap, as revealed here

    https://www.tomshardware.com/news/nvidia-ada-lovelace-successor-in-2025
    hold true to this day?

    Is Blackwell still slated for a 2025 release?

    I'm a little confused.

    Thank you in advance for your time.
    Reply
  • JarredWaltonGPU
    valthuer said:
    @JarredWaltonGPU Sorry to bother you about this, but, in light of these news, when exactly will the RTX-50 series be released?

    Does the initial Nvidia roadmap, as revealed here

    https://www.tomshardware.com/news/nvidia-ada-lovelace-successor-in-2025
    hold true to this day?

    Is Blackwell still slated for a 2025 release?

    I'm a little confused.

    Thank you in advance for your time.
    I believe the original rumors were that consumer Blackwell would be out in 2024–2025, while data center Blackwell would be in 2024. Now, AFAIK, all indications are we won't see consumer Blackwell until 2025. Rubin at present looks more likely to be a 2025 data center only release.

    The only people who know for certain are at Nvidia, and they're not going to say anything officially. I would be very surprised to see Blackwell consumer / RTX 50-series in 2024. It's possible, but I just don't expect it. Process node transitions are slowing down, data center and AI are picking, and so Nvidia has no real need to rush out a new consumer architecture.

    If AMD were to launch a more competitive RDNA 4 next year, then I think Nvidia would move the launch window forward. There's also still the question of what process node consumer Blackwell will use. TSMC N3 / 3N would seem likely, but at the same time, that might be too expensive and we could get 4N again. I'm probably 80–90 percent convinced it will be a "3nm-class" node, but "refined 4N" isn't totally impossible.
    Reply
  • Googulator
    "GR200" sounds like a Grace-Rubin combo chip. (Expect the first letter to change, as in "VR200", if they choose to release a new CPU core "Vera" too.)
    Reply
  • valthuer
    JarredWaltonGPU said:
    I believe the original rumors were that consumer Blackwell would be out in 2024–2025, while data center Blackwell would be in 2024. Now, AFAIK, all indications are we won't see consumer Blackwell until 2025. Rubin at present looks more likely to be a 2025 data center only release.

    The only people who know for certain are at Nvidia, and they're not going to say anything officially. I would be very surprised to see Blackwell consumer / RTX 50-series in 2024. It's possible, but I just don't expect it. Process node transitions are slowing down, data center and AI are picking, and so Nvidia has no real need to rush out a new consumer architecture.

    If AMD were to launch a more competitive RDNA 4 next year, then I think Nvidia would move the launch window forward. There's also still the question of what process node consumer Blackwell will use. TSMC N3 / 3N would seem likely, but at the same time, that might be too expensive and we could get 4N again. I'm probably 80–90 percent convinces it will be a "3nm-class" node, but "refined 4N" isn't totally impossible.

    Yeah, i too would be surprised to see the RTX 50 series out in 2024, considering what beast of a card 4090 is.

    Nvidia, have no competition, so there's no need to rush things.

    Thank you for taking the time to reply.
    Reply