Nvidia Unveils Custom HBM Design With 30% More Bandwidth Than HBM4E
Nvidia introduced NVHBM, a custom high-bandwidth memory technology with 30% more bandwidth than HBM4E. The design moves the memory controller to the base die, improving performance and power efficiency. Nvidia plans to collaborate with suppliers like Samsung and SK Hynix for manufacturing. The industry expects growing demand for custom HBM solutions.
How this was made

The 30-second read
Why it matters
The new memory architecture may improve GPU efficiency, influencing AI infrastructure spending.
Market read
A fresh product launch that could enhance Nvidia's AI GPU competitiveness and affect related memory suppliers.
What to watch
Supply‑chain constraints for HBM production may limit short‑term impact.
Background
Nvidia unveiled NVHBM alongside its Q2 earnings, highlighting a 30% bandwidth gain and up to 15% power savings.
Ticker impact
Nvidia announced its custom NVHBM memory technology, a first‑report product launch with 30% higher bandwidth than HBM4E.
moderate upside over the next weeks as customers evaluate the new memory option
The announcement is fresh, directly from Nvidia, and introduces a tangible performance improvement that may translate into higher sales of AI GPUs.
Market effects
Custom HBM may spur competition among memory suppliers and accelerate AI‑chip differentiation.
Potential upside for US memory makers Samsung and SK Hynix as they may secure new contracts.
Strengthens Nvidia's leadership in AI hardware globally.
Counterpoint
Adoption could be slower if customers face higher costs or integration challenges.
Key entities
- CompanyNvidia
Developer of the NVHBM technology.
- CompanySamsung Electronics
Potential memory supplier for NVHBM.
- CompanySK Hynix
Potential memory supplier for NVHBM.

