The current market is highly focused on the GTC conference. NVIDIA may unveil its next-generation chip codenamed Feynman at the event and make the first public demonstration of products utilizing TSMC’s A16 1.6nm process. This will shift market attention from its Vera Rubin roadmap further into future cycles.

According to Wccftech citing South Korean media Chosun Biz, NVIDIA’s GTC 2026 keynote agenda has “exceeded Vera Rubin,” and this year’s conference may mark Feynman’s first public appearance. GTC 2026 will commence on March 15, with the event returning to San Jose, California, USA.

Jensen Huang had previously stated that his keynote speech would showcase “never-before-revealed” technologies. For investors, such remarks often signal the imminent confirmation of new product schedules and critical supply chain choices, particularly regarding trade-offs between advanced processes and packaging forms.

If Feynman indeed adopts TSMC’s A16, Wccftech believes Nvidia will become the first, and possibly the sole customer, in the early mass production phase of this node, further aligning market expectations for advanced capacity and yield ramping with Nvidia.

Meanwhile, the market is also evaluating whether Feynman will introduce Groq’s LPU units to reduce latency, but this could significantly increase design and manufacturing complexity and impact the production timeline.

The focus of GTC 2026 may shift from Vera Rubin to Feynman

The Chosun Biz report points to a key signal: Nvidia is shifting its narrative focus from Vera Rubin to Feynman at GTC 2026.

Similar to previous methods of announcing new architectures at conferences, Feynman’s presentation may focus on capability overviews, architectural outlines, and production timelines rather than disclosing all details at once.

Currently, technical information about Feynman remains limited, but “looking back one generation” alone is sufficient to prompt the market to reevaluate its product iteration pace in the coming years and its reliance on advanced upstream processes.

TSMC A16, 1.6nm Node: Key Variables in SPR and Early Customer Structures

According to Wccftech, Feynman may become one of the first chips to adopt TSMC’s A16, 1.6nm process. The A16 is described as a significant leap in the semiconductor industry, featuring Super Power Rail (SPR) and is hailed as “the world’s smallest node technology.”.

More notably, the customer structure is significant. Wccftech believes that NVIDIA will be the first major client in the initial mass production phase of the A16 node and “potentially the sole client.”.

Meanwhile, mobile clients or later-stage adopters may implement this standard, as it requires architectural-level modifications. For the market, this implies that the early production capacity utilization and adoption pace of A16 could largely align with NVIDIA’s product strategy.

Speculation on the Groq LPU: Latency Becomes the New Battlefield for GPU Manufacturers

Beyond the generational advancements in manufacturing processes, Feynman also carries another hidden thread: analysis suggests it may be the first to integrate Groq’s LPU hardware stack. The discussion stems from the fact that latency is becoming one of the key metrics GPU manufacturers are focusing on optimizing.

In terms of packaging and integration approaches, the market speculates that NVIDIA may adopt a “hybrid bonding” pathway, incorporating the LPU unit as an on-package option, with its implementation drawing analogies to AMD’s X3D processor.

However, Wccftech also pointed out that this approach would significantly increase the difficulty of design and production, implying that even with a clear direction, the implementation pace may still heavily depend on engineering complexity and manufacturing maturity.

Mass Production Schedule: Production is expected to commence in 2028, with shipments anticipated between 2029 and 2030

In terms of commercialization pace, Wccftech predicts that Feynman production is expected to commence in 2028, with customer shipments likely to occur between 2029 and 2030, depending on NVIDIA’s strategic choices.

This also explains why the GTC 2026 is more likely to be a “forward-looking” release, focusing primarily on architectural outlines and roadmaps to establish expectations for the next-generation platform, followed by gradual fulfillment in mass production and delivery.