Micron Expands Its AI Memory Story Beyond HBM as Analysts Double Their Sights to $1,100

Micron unveils HBM4, 256GB DDR5, and PCIe Gen6 SSD; Raymond James doubles target to $1,100 as AI memory shortage extends through 2028.

Micron Technology is rewriting the AI memory narrative this week at COMPUTEX 2026, pushing far beyond the high-bandwidth memory that has powered its spectacular rally. The stock, which hit a fresh 52-week high of €888 on Monday, has surged more than 80% in the past month alone and now trades at around €898 — nearly ten times the sub-€90 level of a year ago. With a market capitalization that has crossed the trillion-dollar threshold, the company is trying to convince investors that its opportunity runs through the entire memory stack, not just HBM.

That message got a powerful endorsement on June 2 when Raymond James doubled its price target on Micron from $530 to $1,100, maintaining an “Outperform” rating. The analysts pointed to a global shortage of high-performance memory that they expect to persist for years. Goldman Sachs has even extended its supply-constrained view on AI infrastructure memory through 2028 — an unusually long visibility horizon that underpins revenue expectations.

The product lineup on display in Taipei is designed to turn that thesis into something tangible. Micron unveiled its HBM4 36GB 12H stack, claiming it can boost inference throughput for large language models by a factor of 2.6. Alongside it came the 256 GB SOCAMM2, which the company says uses roughly a third of the power and physical footprint of traditional RDIMMs. A new 256 GB DDR5 RDIMM based on 1? technology also made its debut, with top speeds of up to 9,200 megatransfers per second — 40% faster than current production modules while consuming over 40% less operating power.

The strategic significance of these products was underscored a day before the target hike. Nvidia CEO Jensen Huang confirmed on June 1 that the upcoming Vera-Rubin system will integrate memory from Micron, SK Hynix and Samsung. For Micron, that is a direct entry ticket into the core of AI infrastructure — no longer a supplier on the periphery.

Yet the company is working to broaden the story well beyond data-center HBM. On the SSD front, Micron highlighted the 9650 as the first commercially available PCIe Gen6 drive for AI inference and training. The 6600 ION, available in capacities up to 245 TB, can reportedly cut required rack space by 82% and halve power consumption compared to HDD-based setups.

Edge AI is another expansion vector. Micron is pushing higher DRAM densities into PCs, smartphones, vehicles and embedded systems, where models and agents can run directly on the device. LPCAMM2 modules with LPDDR5X reach data rates up to 9,600 MT/s, while GDDR7 delivers up to 1.5 TB/s of system bandwidth — a 60% jump over GDDR6 with 33% better AI inference performance. Automotive UFS 4.1 storage, rated at up to 4.2 GB/s and functional safety support, targets advanced driver-assistance systems and in-vehicle AI.

The financials already show the momentum. In Micron's second fiscal quarter, ended February 26, 2026, revenue hit $23.86 billion — up from $13.64 billion in the prior quarter and $8.05 billion a year earlier. GAAP net income came in at $13.79 billion, with operating cash flow of $11.90 billion. For the third quarter, which the company will report on June 24, consensus estimates call for revenue of around $33.5 billion and adjusted earnings per share of $19.15. That would follow a Q2 beat when EPS of $12.20 crushed the $9.19 consensus.

Institutional investors have taken notice. BNP Paribas boosted its Micron position by over 1,000% in the fourth quarter to roughly 37,600 shares, while COFG Advisors and Beto Financial Group opened new stakes. Not every signal points skyward, however. CEO Sanjay Mehrotra has trimmed his own holdings, and some market watchers draw uncomfortable parallels to the semiconductor euphoria of 1999.

Proponents of the current cycle counter with a hard-number rebuttal: hyperscalers such as Alphabet and Amazon are collectively investing more than $700 billion in AI infrastructure, backed by firm orders and long-term contracts rather than speculative excess. The quarterly report on June 24 will provide the next real-time test of whether Micron's broadened product story can keep pace with a stock that has already priced in so much.