Nvidia's NVHBM Brings 30% More Bandwidth to NVLink Fusion; Amazon Annapurna First Partner
nordicinst · x · 2026-08-27
NVIDIA officially expanded NVLink Fusion with NVHBM, a custom high-bandwidth memory technology, with Amazon's Annapurna Labs as the first collaborator.
Key points: traditional HBM places the memory controller on the XPU die, consuming compute silicon. NVHBM integrates NVIDIA's custom memory controller into the HBM base die (same tech as future NVIDIA GPUs), delivering versus standard HBM4E:
- Up to 30% greater memory bandwidth
- 15% lower HBM power consumption
- Up to 25% more area freed on the XPU compute die
NVIDIA will standardize NVHBM across multiple memory providers, cutting qualification effort so NVLink Fusion customers can bring semi-custom AI chips to market faster.
Related event: NVIDIA Unveils NVHBM Custom High-Bandwidth Memory(2 posts)→
More from Infra
- NVIDIA releases Nemotron 3.5 Bolt 74B MoE that fits in 24GB VRAM at 2-bit — NerdyRodent · 2026-10-16
- Analyst: 80% of neoclouds may not survive, but the demand for their GPU capacity is real — BenBajarin · 2026-10-16
- Apple and Meta partner to run Muse Spark model on linked Mac Studios — pdamodaran · 2026-10-16
- Nvidia's developer base hits 11 million, up ~3.7x from 3 million in 2021 — Beth_Kindig · 2026-10-16
- Muse Spark Open runs locally on a Mac M3 Ultra via llama.cpp: frontier model on desktop — pcuenq · 2026-10-16
- Muse-Spark open-weights drop: 816B-A42B MoE model gets 359GB MLX 3-bit build — ns123abc · 2026-10-16