NVIDIA Unveils NVHBM: 30% Higher Bandwidth and 15% Lower Power
zephyr_z9 · x · 2026-08-27
NVIDIA expands NVLink Fusion with NVHBM, a next-gen high-bandwidth memory technology. By integrating the memory controller into the HBM base die instead of the XPU, NVHBM saves valuable silicon area. Compared to standard HBM4E, it delivers up to 30% greater memory bandwidth, reduces HBM power consumption by 15%, and frees up to 25% more area on the XPU compute die. Amazon's Annapurna Labs is the first partner to collaborate on NVHBM to enhance AI workload performance on AWS custom silicon.
Related event: NVIDIA Unveils NVHBM Custom Memory, Boosting Bandwidth 30%(2 posts)→
More from Infra
- Jensen prices the watt: GPU revenue per GW to hit $64B — sudoraohacker · 2026-08-27
- Colibrì Engine Update: Runs 2.8T Param Models, Boosts Speed via Expert Caching — solyarisoftware · 2026-08-27
- Jensen Huang: Data Center Investment Payback Period Under One Year — zephyr_z9 · 2026-08-27
- vLLM promotes TPU to first-class backend with new unified support — vllm_project · 2026-08-27
- 320B Model Runs on Mac: OrcaSAQ Quantization Brings GLM-5.3 to Apple Silicon — alejandroll10 · 2026-08-27
- GLM 5.3 Flash hits RunInfra day one: 254 tok/s, 1M context, $0.10/1M input — alejandroll10 · 2026-08-27