KMEM vs DRAM vs HBMX vs DRMP: Four Memory Bets

Overview

These four securities track the memory semiconductor sector—DRAM, NAND flash, and high-bandwidth memory producers plus their supply chain—but differ sharply in structure and purpose. DRAM and KMEM are passive or index-tracking ETFs focused on capital appreciation; DRMP layers a weekly income strategy using put credit spreads on memory stocks; HBMX concentrates on fewer holdings within the ecosystem. All four are recent launches (June 2026), making them untested through a full market cycle.

How they differ

The biggest distinction is income strategy. DRMP is the only one designed to generate distributions, using an active put credit spread overlay on memory equities to target a 27.10% yield paid weekly. That's fundamentally different from DRAM, KMEM, and HBMX, which distribute nothing or only annual capital gains. The second key difference is concentration. HBMX is explicitly concentrated on fewer holdings to maximize exposure to high-bandwidth memory and advanced packaging; DRAM and KMEM cast a wider net across the memory ecosystem; DRMP sits in between, holding memory-stack equities while running options strategies on top. Third is fee and cost structure. DRAM and KMEM charge 0.65% in expense ratio with $23.7B and $26.8M in AUM respectively; DRMP and HBMX both charge 0.95%, but DRMP's $6.26M AUM is tiny and may face scaling pressure, while HBMX's $27.8M is also modest for an actively managed fund.

Who each is best for

DRAM: Fits investors seeking broad memory semiconductor exposure with no income requirement and a lower cost structure; the $23.7B AUM suggests established liquidity and track record.

DRMP: Designed for income-focused traders willing to accept weekly distributions, options-strategy risk, and the tradeoff of potential NAV erosion in exchange for high current yield.

HBMX: Fits investors who believe high-bandwidth memory and advanced packaging represent the next wave of semiconductor growth and prefer concentrated, actively managed selection over a passive basket.

KMEM: Matches investors looking for straightforward memory semiconductor exposure at a low 0.65% expense ratio without an active manager or options overlay.

Key risks to know

• NAV erosion at extreme distribution yields: DRMP's 27.10% weekly distribution rate far exceeds the underlying equity sector's typical return potential, meaning the fund is likely returning significant capital or relying on put spread premium that may not sustain through market stress or volatility contraction.

• Options and spread strategy risk in DRMP: A systematic put credit spread strategy benefits from declining realized volatility and stable memory stock prices. A sharp selloff in semiconductor memory or a spike in implied volatility could force large losses on short puts while the underlying equities also decline.

• Concentration and tracking error in HBMX: An actively managed, concentrated fund explicitly targeting fewer holdings will deviate significantly from the broader memory market, introducing single-stock and sector rotation risk that a passive index does not.

• Extreme recency and size: All four funds launched within four weeks (June 2, 2026 to June 30, 2026). DRMP and HBMX have minimal AUM ($6.26M and $27.8M), raising questions about liquidity depth, ability to attract assets, and whether the strategies will persist if capital doesn't grow.

• Memory semiconductor cyclicality: The entire memory ecosystem is highly cyclical and correlated with AI capex spending and DRAM/NAND pricing cycles. A demand correction or inventory glut can hit all four funds simultaneously, regardless of structure.

Bottom line

If you want simple, passive memory semiconductor exposure with established scale, DRAM stands out; KMEM offers the same approach at identical cost with smaller AUM. If you prioritize weekly income and can tolerate options risk and NAV volatility, DRMP's 27.10% yield is what you're paying for—though verify that the put spread strategy sustains outside a rising market. HBMX fits active believers in high-bandwidth memory as a secular shift, but concentration introduces idiosyncratic risk that passive funds avoid. Past performance does not predict future results, and all four are untested in a full cycle.