Broadcom is among the major customers securing HBM supply for its custom AI accelerators, indicating broad-based demand for memory across merchant and custom silicon.

Google is actively managing HBM allocations for its TPU accelerator program, reflecting the same supply chain pressures facing Nvidia and custom silicon vendors.

Moving to 4-high HBM more than doubles harvestable cubes, shifting the bottleneck from HBM wafers to TSMC leading-edge logic wafers and advanced packaging capacity needed to co-pa…

Memory vendors are printing massive profits but cannot respond to demand signals quickly because adding wafer capacity requires building clean rooms, filling them with equipment,…

Samsung's memory business is doing well alongside SK Hynix and Micron, but all three face identical long lead times for capacity expansion driven by clean room construction and AS…

Micron is benefiting from tight DRAM/HBM markets but faces the same multi-year capacity expansion constraints as SK Hynix and Samsung, limiting ability to meet AI-driven demand.

Memory capacity growth is gated by ASML's ability to produce EUV lithography tools, which themselves require specialized suppliers (e.g., ultra-smooth mirrors) with long lead time…

Meta is actively requesting semi-custom accelerator variants (e.g., MI450 with 8-high HBM) to avoid overprovisioning expensive memory capacity it doesn't need, setting a precedent…

AMD is already providing custom memory configurations (8-high vs 12-high HBM on MI450) for hyperscalers like Meta, signaling a broader industry shift toward multiple SKUs per acce…