Marvell expects its revenue to grow tenfold in five years to $80B, driven by custom chips it builds for hyperscalers. Those companies design their own chips to reduce reliance on Nvidia and cut the long-term cost of off-the-shelf parts. The custom segment had no revenue two years ago and is expected to bring in $30B by 2031. x.com 

Marvell is a 2030 stock: the market sold it 32% off its high because the payoff starts in 2029. Data center is 79% of revenue and growing 46%. Management raised its fiscal 2028 target to $18B, said custom silicon more than doubles next year, and in August signed a Google TPU agreement with up to $120B of revenue potential over six years. x.com 

Marvell's ambitious long-range outlook is encouraging about hyperscaler and AI-lab spending plans, but hitting it depends on AI revenue and token consumption soaring through decade's end. Even if its FY28 estimates rise again, Marvell trades at a healthy premium to Nvidia and Broadcom, especially on a GAAP basis, and those two are growing fast too. x.com 

Marvell's $70-90 billion FY31 revenue target looks credible: COO Chris Koopmans said key foundry, substrate and packaging suppliers get five-year forecasts and had already seen it before investor day. A $500 price target still looks too aggressive for now; Marvell must show it can grow its networking and photonics segment to deserve a safe price increase. x.com  x.com 

Marvell can become the next $1 trillion company, which makes the investing case simple. Jensen Huang was cited as saying so, and even if that comment actually referred to AMD, Marvell can still reach the mark. x.com  x.com 

Marvell's $1 trillion valuation, which Jensen Huang predicted, is a slogan rather than a valuation. At its FY2028 targets of $16.5B revenue and a high-30s operating margin, $1T means about 60x sales and 180x adjusted earnings; at 30x earnings it would need roughly $95B of revenue, essentially the entire $94B 2028 data center market Marvell estimates. GAAP net income was just $34.5M in Q1, and dilution adds pressure. Operating leverage from 45% growth is the bull case. x.com 

Marvell's valuation still reflects old cyclical models, overlooking an AI-driven structural shift that leaves room for a re-rating. Marvell focuses on AI networking and custom silicon, with data center above 70% of revenue and a lead in photonics interconnects, while long-term agreements secure demand and reduce cyclicality. Its EPS growth depends on execution. x.com 

Marvell is being repriced as a core AI infrastructure holding after record results, with the bull case shifting from custom silicon toward networking and optics. Fresh June targets run $321-$385, led by KeyBanc's Street-high $385 on data center networking, scale-up and the $3.25 billion Celestial AI optical acquisition; the $236 average target is lagging, not bearish. The catch: the stock already trades above that average, and KeyBanc calls the custom XPU business less durable long term. x.com 

Marvell's strong quarter was already priced in, so a sell-the-news decline is likelier. Q2 revenue rose 37% to $2.739B, data center revenue 46%, and Q3 guidance of $3.15B beat estimates; management sees custom silicon revenue more than doubling in FY28 and a path to about $18B of FY28 revenue. But the Google partnership and AI growth were largely traded ahead, and the stock never broke its rising channel's top. Long term, it is still a good company to add to at better prices before the midterms. … Show more

Marvell's connectivity and custom AI silicon support raised revenue targets of about $12bn for FY27 and $18bn for FY28, even as shares fell 7.8% after hours on high expectations. F2Q revenue of $2.74bn rose 37% year over year and F3Q guidance of $3.15bn implies 50% growth. Management guides custom ASIC growth above 100% in FY28, with a path to $16bn+ by FY29 versus its ~$10bn, helped by a multi-year Google deal. AI clusters needing ~10x more bandwidth favor 1.6T optics. Buy, $290 target. x.com  … Show more