NVIDIA's GTC 2026 keynote plans are looking far beyond Vera Rubin, according to a new report, which says the company could showcase next-gen Feynman chips.
NVIDIA's Feynman Chips Might Also Mark the First Instance of the Adoption of Groq's LPU Units
We already know that Team Green plans to set the tone for the next decade of computing at this year's GTC, and, according to Jensen himself, the world will see technology "never unveiled before" in his keynote. While we did speculate back then that the showcase would be dedicated to Feynman, it appears the Korean media outlet Chosun Biz agrees with us, as their latest report indicates that this year's GTC will mark the first public debut of Feynman.
Details on the Feynman chips are currently slim, but we do know that it will mark the first adoption of TSMC's A16 (1.6nm) chips, which are also a major leap in the realm of semiconductors, featuring Super Power Rail (SPR) and the world's smallest node technology. NVIDIA is set to be TSMC's first, and probably the only customer for the A16 node during the initial HVM phase, given that it is claimed that mobile customers might adopt the chip standard later, as it requires architectural revamps. However, Feynman showcases at GTC 2026 also have another interesting angle.
We believe that Feynman would mark the first integration of Groq's LPU hardware stack, as latency has become a major focus for GPU manufacturers. In a previous report, we discussed how Team Green could use a hybrid bonding approach with LPU units, and the implementation might be similar to what we witnessed with AMD's X3D processors. While the exact architectural designs aren't specific, it only makes sense for LPUs to be an on-package option with Feynman, but doing so would make designing and production much more difficult.
NVIDIA's Feynman showcase could be similar to what we saw with Vera Rubin back when the GB200s were under full production, featuring a rundown of capabilities, a general architecture overview, and mass-production timelines. As for when Feynman will debut, we anticipate production to commence by 2028, which could push customer shipments to 2029-2030, depending on NVIDIA's strategy. GTC 2026 starts on March 15, and the event returns to San Jose, California for this year.
About the author: Muhammad Zuhair is a hardware and technology reporter for Wccftech, specializing in the semiconductor industry and the complex interplay between technology, manufacturing, and geopolitics. His coverage focuses on the corporate strategies and technological roadmaps of industry giants like TSMC, NVIDIA, Samsung, and Intel. Zuhair's expertise lies in deconstructing complex topics such as fabrication nodes (e.g., 2nm process), the economic impact of policies like the CHIPS Act, and the strategic development of AI infrastructure from NVIDIA, AMD and Intel.
Follow Wccftech on Google to get more of our news coverage in your feeds.
