Small Cap Snipa
AI & Data-Center Equity Research, Power Infrastructure & Cloud Compute
Premium-model inference revenue could support much higher compute pricing for AI infrastructure providers. Evercore models over $120M/MW in potential lab revenue while labs pay roughly $30-50M/MW for compute. With power and GPUs scarce, SpaceX AI, Nebius, IREN and CoreWeave can capture the biggest uplift over historical pricing if they deliver contracted high-density capacity. x.com x.comÂ
Networking's growing share of data-center spending and its role in high-end compute performance, per J.P. Morgan, support a bullish view on Arista, Coherent, Lumentum, Credo, Amphenol and Corning. The bank raised its top-four U.S. cloud providers' data-center capex growth forecasts to 85% for 2026 and 70% for 2027, implying annual spending increases of more than $270B and $400B+, respectively. x.comÂ
AI agents' productivity over human labor should lift GPU demand and rates for neoclouds. The bottom looks in for CoreWeave, Nebius, Boost Run, Sharon AI and IREN. Gavin Baker says token prices can fall as revenue per GPU rises; paraphrasing Elon, even $38 per GPU-hour can pay off. CoreWeave's five-year fixed contracts may lag current rates; Boost Run has less capacity locked in. If SpaceX takes more growth, slower cash burn could rerate neoclouds, though smaller customers may pay less. x.com … Show more
SpaceX can capture more upside from AI compute scarcity by selling capacity near spot prices instead of locking in today's rates. Potential Anthropic and Google commitments total about $151B, with 90-day termination rights, while bare-metal pricing has risen from $10-15M to nearly $50M in annual revenue per MW. Elon Musk says contract payback could be as short as six months in 2027. x.com x.comÂ
Hardware suppliers and VC keep financing cash-hungry labs and neoclouds, making compute rental prices at CoreWeave, Nebius and IREN look artificially supported. Lab revenue growth is finally plateauing at 800% year over year after roughly tenfold equity gains across the stack over the last decade. The buildout is also crowding out the economy's appetite for debt. x.com x.com x.com x.com x.com x.comÂ
SpaceX's AI compute buildout has more evidence of demand from big customers. Per Sawyer Merritt, Anthropic's IPO filing says it agreed to spend up to $120.5 billion on SpaceXAI compute through 2029, with payments of about $3.1 billion a month once all capacity is delivered, expected in December 2026. Either side may cancel on 90 days' notice after initial terms end in January and March 2027. x.com x.comÂ
A rising value of compute per watt should benefit AI cloud, chip, connectivity and power suppliers. Elon Musk says that value will rise over time. Nebius has Microsoft and Meta deals; Marvell benefits from custom silicon demand; Credo connects GPUs; Vistra has long-term nuclear supply deals with Amazon and Meta; Nokia's Nvidia partnership could help it capture data-center networking growth. x.com x.comÂ
SpaceX's AI-cloud ambitions face a severe two-sided constraint: finding 12 GW of Rubin capacity in one year looks implausible, and even that buildout would need customers willing to rent roughly $50B of compute annually. The scale of both the supply and demand challenges makes the plan highly skeptical. x.comÂ
Improving AI models could raise the value of compute per watt rather than erode neocloud economics. SpaceX's Elon Musk argues that better models yield more useful intelligence per watt, points to older H100s renting for more than when new, and expects a lasting shortage of advanced compute. That statement is heavily bullish for neoclouds such as Nebius, though the mechanism is Musk's forecast. x.comÂ
Accelerating compute demand could drive a V-shaped recovery in beaten-down neocloud stocks if SpaceX earnings confirm it. Nebius, CoreWeave and IREN are on watch. GPU spot rates rose rapidly: H100s $3.34/hour, H200s $4.93, B200s $8.13. Nebius had $582M Q2 revenue, nearly 6x YoY; CoreWeave a $129B backlog early in Q3; SpaceX 1.4 GW compute capacity last quarter, up from 400 MW a year ago. Whether SpaceX converts demand into meaningful revenue and EPS beats remains the key question. x.comÂ
SpaceX's 13-20GW 2027 plan would crash compute rates if achieved, and Nvidia cannot supply the chips. Even 10GW gross/8GW IT needs 350,000 GPUs and $30-32B monthly; 20GW costs $400-500B against $750B in global AI spend. A 28% share release tomorrow and Musk's cross-company overpayments keep SpaceX hands-off. Yet 400MW+ monthly deployment could yield $225B revenue and $100B profit in 2027, worth $2T at 20x; Elon firms rarely trade at only 20x. x.com x.com x.com x.com x.comÂ
Nebius' software efficiency and a possible end to heavy AI financing support a re-rating. Per @DrTomsLens, its engineers got about 43% more video output from the same GPUs, cutting compute cost per second by 30%; newer hardware alone is not enough without capable software teams. Oracle expects to fund growth from free cash flow in FY2029, suggesting capital-heavy AI providers may eventually outgrow debt and dilution. Google Cloud reported 103% revenue growth amid strong AI infrastructure demand. … Show more
