Iren vs. Nebius: Which Neocloud Provider Is the Better AI Stock?

The article compares Iren and Nebius as AI neocloud leaders, highlighting Nebius's stronger recent dealmaking, including two Meta contracts totaling $27 billion over five years, versus Iren's last major deal in November. Iren’s long-term upside is tied to 4.5 GW of total portfolio potential, including a 1.6 GW Oklahoma site and the upcoming 1.4 GW Sweetwater 1 project, but only 810 MW is currently operational. The piece argues Nebius can charge more per megawatt thanks to its software stack, while Iren may offer greater revenue expansion potential once its secured power comes online.

Analysis

NBIS has the cleaner near-term narrative because it is proving it can convert hyperscaler demand into repeated contract wins, which matters more than raw headline backlog in this phase of the cycle. The market is likely to keep paying up for any operator that demonstrates pricing power per MW and faster customer diversification, because that reduces the risk that these names become simple utility-like capacity trades rather than software-enabled infrastructure platforms.

IREN is the more interesting medium-term setup. The market is underestimating how much embedded optionality sits in its unenergized pipeline: once a large block of power flips from "future capacity" to live capacity, the revenue base can re-rate very quickly without needing a new bull case. That creates a classic second-order effect: a single execution milestone can pull forward multiple years of ARR recognition and force incremental demand from peers, suppliers, and financing partners who want exposure to the remaining buildout.

The key risk is that this remains a timing stock, not a pure fundamentals compounder yet. If energization slips by even one quarter, the valuation multiple can compress faster than the fundamental damage, because investors are currently paying for execution certainty rather than long-dated optionality. Conversely, if IREN lands one more marquee agreement alongside successful commissioning, the stock could gap materially as the market starts capitalizing the pipeline instead of waiting for it.

Consensus is too focused on who has the most gigawatts and not enough on who can monetize each gigawatt most efficiently. NBIS likely retains a pricing-quality premium, but IREN may have the superior convexity because scale and available power convert into revenue step-ups much faster once deployment risk is removed. In short, NBIS is the better quality near-term asset; IREN is the higher-beta upside call on execution.

Key Decisions for Investors

• Long NBIS / short IREN as a short-duration relative-value trade into IREN earnings: thesis is that NBIS continues to monetize scarcity at a higher $/MW while IREN remains execution-risk heavy; target 5-10% spread compression over 1-2 months, exit if IREN announces a new hyperscaler deal or meaningful energization progress.
• Buy IREN on weakness only after the next earnings print, using a layered entry over 2-4 weeks: the setup improves if management confirms commissioning is on track, with upside convexity if a new contract lands; risk is a one-quarter delay that can knock 10-20% off sentiment quickly.
• Use call spreads on IREN for a catalyst trade into the next 1-2 quarters: preferred structure is bullish but defined-risk, since any single deal or capacity milestone could rerate the stock sharply, while the downside is capped if execution slips.
• Stay overweight NBIS versus other neocloud proxies for the next 3-6 months: it is the cleaner beneficiary of hyperscaler demand with better evidence of pricing power, making it the higher-confidence hold even if the upside is less explosive than IREN.
• Avoid chasing both names after headline deal announcements; wait for either confirmed energization or disclosed MW economics, because the market tends to overpay for backlog headlines before demanding proof of deployment.