On June 16th, the "AI End-of-Day Report" published by Citrini Research stated that AMD and Apple are currently advancing solutions that replace DRAM with flash memory in AI products. Among them, AMD's acquisition of MEXT optimizes flash memory to approach the performance of DRAM, thereby reducing memory costs in AI data centers; Apple achieves similar optimization on the device side through "LLM in a flash" technology.

According to Citrini's latest research report, the high demand for KV cache in AI inference and the "memory tax" pressure, with HBM already accounting for 25% of DRAM capacity. Meanwhile, flash memory costs only 1/55 of DRAM, and through controller optimization, NAND stacking, and cell mode adjustments, it can provide an alternative for capacity and bandwidth in edge AI.

The above report provides theoretical support for the recent continuous surge in storage stocks represented by SanDisk. #我的Gate交易时刻