CoreWeave vs Nebius Group

The neo-cloud pair. These are two of the fastest-growing companies ever scored by BlitzLab — and both land in the 50s. This is the clearest illustration in the index of a single rule: speed bought with debt gets graded twice, once as fuel and once as a bill.

The success line is 65. Demerits (section D) are subtracted; a demerit total of 15 or more flags a company for BlitzFailing watch. Scores as of the 2026 Q4 review (October 21, 2026).

Speed — the fastest pair in the index

Between them they hold the two highest speed scores in this set: CoreWeave 21, Nebius 20. CoreWeave’s revenue hit $5.13B in FY2025 (+168%) and its trailing twelve months are about $7.6B, with FY2026 guidance raised to $12.4–13.2B; revenue per employee of roughly $3.5M is the highest productivity figure we have measured. Nebius is growing from a smaller base at an even steeper angle: FY2025 revenue of $529.8M (+479%), ARR up from $1.9B in March to $3.0B by June, and FY2026 ARR guidance of $7–9B. Both pass the market-size gate easily — CoreWeave’s $104B backlog alone exceeds its ten-year bar, and Nebius holds a $40B-plus backlog against a $13.6B bar (B3: 10 each).

Moat — capacity is not a network

Neither company has a true network effect, and the scorecards say so: B1 is 8 and 7, earned from switching costs, multi-product attach (90% of CoreWeave customers use two or more products) and privileged access to NVIDIA’s newest chips rather than from any user-to-user flywheel. CoreWeave deploys faster (B2: 13 vs 12 — 850MW live heading to a 1.7GW target, against Nebius at 170MW live heading for 800MW–1GW). Nebius claws points back on margin quality (B4: 7 vs 6): its adjusted EBITDA has already turned positive and its margin turnaround is running ahead of CoreWeave’s, where GAAP gross margin of about 24% after depreciation and a −25% net margin show how much of the unit economics interest and depreciation consume. Nebius also scores the stronger behavior lock-in (C2: 4 vs 3) — Microsoft’s five-year $19.4B contract and Meta’s $12B committed deal are structured with heavy prepayments; about 70% of its Q2 contracts included them, which is why we call it the prepaid blitz.

Demerits — the balance sheet decides

The four-point gap is, almost entirely, a demerit gap: −13 against −8. CoreWeave’s D1 is the heaviest single demerit in this set at −8: total debt above $35B is roughly 4.6× trailing revenue, quarterly interest of $640M runs at about five times adjusted operating income, and the FY2026 capex guide of $35–39B is nearly three times revenue guidance. Nebius is hardly unlevered — pro-forma debt around $13.5B is roughly ten times its smaller trailing revenue, scored −7 — but its interest burden relative to earnings is lighter and its buildout is substantially pre-funded by customers. CoreWeave pays a further −3 for the securities class actions filed in early 2026 alleging overstated demand capacity and concealed construction delays, and −2 for $2.3B of founder selling after the lock-up expired. Both companies may yet blitz their way out; the index simply records that, today, Nebius’ speed is financed in a way that leaves more of the score intact.

How to read this

Both companies passed the same two gates before scoring: real product-market traction (G1), and a ten-year market ceiling of at least ten times current revenue (G2). The Blitz Index then grades speed (25), moat (60) and creative destruction (15), and subtracts BlitzFailing demerits (up to −30) for debt, revenue quality, governance and regulatory cost. The model is frozen at v1.0; past scores are never silently rewritten. Our own pre-registered tests found the score measures business quality — it does not predict stock returns.