NVIDIA unveils next-gen Feynman GPU In GTC 2025 roadmap, should use HBM5 memory in 2028

NVIDIA has just announced that its next-gen Feynman GPU architecture will succeed Rubin and Blackwell, and will debut with ''next-gen HBM" memory in 2028.

2

At its GPU Technology Conference (GTC 2025) event, NVIDIA unveiled its latest AI GPU architecture roadmap, extending beyond the just-detailed Rubin and Rubin AI GPUs. NVIDIA has now confirmed that its next-gen AI GPU will be codenamed Feynman, after renowned American theoretical physicist, Richard Phillips Feynman, who is known for his work on quantum mechanics

On his Wikipedia page, Feynman is explained as "an American theoretical physicist. He is best known for his work in the path integral formulation of quantum mechanics, the theory of quantum electrodynamics, the physics of the superfluidity of supercooled liquid helium, and in particle physics, for which he proposed the parton model".

NVIDIA didn't detail which next-gen HBM memory it would be using, but we know that its current GB200 and new GB300 AI GPUs use HBM3E memory, and its next-gen Rubin AI GPUs will use HBM4... so we could expect HBM4E memory or next-gen HBM5 memory. NVIDIA's new Feynman AI GPUs will use the same Vera CPUs that will be used on its upcoming Rubin and Rubin AI GPUs, too.

• Read more: NVIDIA GB300 'Blackwell Ultra' AI GPU: 288GB HBM3E, 50% faster than GB200

NVIDIA has used an "Ultra" moniker on Blackwell with Blackwell Ultra, as well as its next-gen Rubin with Rubin Ultra, so we should expect Feynman to be joined by Feynman Ultra.. NVIDIA's next-gen Feynman should launch in 2028, while Feynman Ultra be released in 2030 if it becomes a thing.