NuAxiSemiconductor trade intelligence

NVIDIA Shares "Rubin" GPU Deep-Dive and Die Annotation

What happened

NVIDIA unveiled the Rubin GPU architecture featuring 224 SMs, 288GB HBM4 memory, and NVLink 6. The design targets massive data center compute scaling. 关注 - Technical roadmap without immediate supply chain disruption.

Impact and signal

Low impactImpact score: 10/100

Affected components and suppliers

HBM4GPUNVLinkTensor Cores

NVIDIA

roadmaphbm4nvlink

Source and time

Source: TechPowerUp · 7/21/2026 · View original source

Open the NuAxi live feed →

Real-time pricing, supply signals and AI analysis for semiconductor traders

NVIDIA Cuts HBM4 Speed and NVLink Bandwidth Targets for "Vera Rubin"TechPowerUp · 10/26/2026
[News] NVIDIA’s Memory Shake-Up: Rubin Ultra Rethinks 12-Hi HBM4E for 8-Hi HBM4, Rubin GPU Bandwidth Cut 13%TrendForce · 10/26/2026
📈 #NVIDIA RTX 50 graphics cards could rise by up to 30% in Korea from August as #TSMC 4nm wafer and #GDDR7 memory costs increase. RTX 5090 (32GB) p...TrendForce Official (Market Data) · 8/4/2026
[News] Samsung, SK hynix’s HBM4 Push Puts HBM, General Memory Pricing in the Spotlight for 2H EarningsTrendForce · 8/12/2026
🚨 Amid the AI boom, securing enough GPUs and CPUs has become increasingly difficult—and even renting them isn’t getting any easier. In China, ren...TrendForce Official (Market Data) · 4/17/2026
Anthropic reportedly commits US$45 billion to Nscale for AI computing capacityDigiTimes Asia · 8/26/2026

NVIDIA Shares "Rubin" GPU Deep-Dive and Die Annotation

What happened

Impact and signal

Affected components and suppliers

Source and time

Related briefings