Analysis
Nscale's pending Nasdaq listing has been reported at valuations anywhere from $25 billion to $35 billion, per figures tracked on Pulse's stats page -- a roughly $10 billion spread for a company nearing an actual roadshow, not an early-stage private round where wide ranges are normal.
## Why the range hasn't narrowed That gap reflects a real disagreement over how to value capacity that's contracted but not yet delivered. Nscale's case leans heavily on its Microsoft and Anthropic contracts, and its ratio of promised-to-online capacity is reportedly facing the same investor skepticism that sank Firmus Grid's $30 billion IPO after it delivered only 46 of a promised 912 megawatts.
Nscale has built out considerably more capacity than Firmus Grid did, which is likely why its range tops out above Firmus Grid's failed ask rather than collapsing the same way. But the fact that credible reports still disagree by $10 billion this close to pricing suggests bankers haven't converged on a single framework for valuing neocloud backlogs -- whether that's a multiple of signed contract value, of megawatts delivered, or of megawatts merely under construction. That concentration risk is already showing cracks: Microsoft has stepped back from at least one Nscale UK project over grid delays, even as it remains a strategic partner tied to Nscale's broader buildout elsewhere.
“## Why the range hasn't narrowed That gap reflects a real disagreement over how to value capacity that's contracted but not yet delivered.”
For Nscale, narrowing that range before the roadshow matters more than which number wins: a listing that prices inside a tightly argued range signals conviction, while one that prices at either extreme of a $10 billion spread invites exactly the kind of second-guessing that followed Firmus Grid out the door.
Other neoclouds racing toward a listing face the same test. DayOne's roughly $20 billion target has stayed comparatively stable because its capacity growth is already delivered and auditable, not projected -- exactly the kind of evidence Nscale's bankers will need to collapse their own $10 billion spread before investors do it for them at the low end.
The lesson for anyone pricing the next neocloud listing is the same one Firmus Grid's investors already learned the hard way: a valuation built on signed contracts and construction timelines is a story, and a valuation built on megawatts actually switched on is a balance sheet. Until Nscale's bankers can point to more of the latter, expect the range to stay wide.
