NVIDIA Prototypes Next-Gen Feynman GPU Architecture

NVIDIA is developing its Feynman GPU using TSMC's 1.6nm node to deliver over 1,000 TB/s bandwidth via advanced 3D chiplet packaging.

NVIDIA is currently prototyping its next-generation Feynman GPU microarchitecture, which is set to replace the Rubin and Rubin Ultra designs. The company is leveraging TSMC's 1.6nm A16 node for production, with mass manufacturing scheduled for late 2028.

The Feynman architecture will utilize backside power delivery and advanced 3D chiplets via SoIC packaging. These designs integrate CoWoS-L 2.5D technology and panel-level packaging, alongside customized HBM4E memory featuring logic-capable base dies.

This shift aims for a massive performance leap, targeting over 1,000 TB/s in total system bandwidth using COUPE optical interconnects. This doubles the 520 TB/s capacity of Rubin Ultra, significantly advancing the Blackwell NVL72 rack's 130 TB/s baseline.


More articles

Intel to Support Raptor Lake CPUs for Years

Gigabyte Launches New B450M D3HP Motherboards

Snapdragon 8 Elite Gen 6 Pro Benchmark Leaks

DapuStor Unveils World's First 512 TB PCIe 5.0 SSD

AMD Launches $5 Billion Senior Debt Offering

Geometric Future Debuts Model 3 Case and Cooling

Researcher Unveils Skitter Creek Exploit for AMD

CoreWeave Secures Nvidia A100 Contract to 2029

Intel Nova Lake-S Cache Details Leaked

KTC Unveils M27P6S 4K QD-Mini-LED Gaming Monitor

Qualcomm Unveils Snapdragon C for Budget Laptops

Kioxia and SanDisk Unveil 230-Layer BiCS9 QLC NAND

BIOSTAR Launches EdgeComp MT-N150 Edge System

Lenovo Project Swan ThinkBook Leak Reveals Rollable OLED

Beelink Launches ME Pro NAS with Ryzen AI 9 HX 470

Google Unveils Pixel 11 Series with Android 17

TSMC Achieves 98% Yield for AI CoWoS Packaging

Newegg Offers $658 AMD Ryzen 9 9900X Hardware Bundle

Qualcomm Unveils Snapdragon C for Entry-Level PCs

ZOTAC Unveils RTX 5080 20th Anniversary Edition