Boost Run (BRUN): what it does, the $940M backlog, and the financing gap

The standard $BRUN pitch is the one every neocloud runs: a giant backlog, triple-digit growth, and a discount to the bigger names. Boost Run has all three. The question its roughly $2.2B price tag skates past is the only one that matters for a company three weeks old: can it actually finance the backlog?

The honest frame: Boost Run is a micro-cap "neocloud" that rents AI compute, and its equity story is a financing-gap bet, not a revenue bet. It is sitting on a $940M contracted backlog and a $1.44B GPU hardware bill to $DELL, against about $9.7M of cash plus the de-SPAC trust. This piece walks through what a neocloud does, how Boost Run makes money, where it sits versus $CRWV and $NBIS, and why the balance sheet is the whole trade.

Why it matters now

Boost Run began trading on Nasdaq on May 11, 2026, after a de-SPAC merger with Willow Lane Acquisition Corp closed on May 8. It listed with a $940M backlog and a forecast of roughly 250% revenue growth for 2026, and the stock promptly did what thinly-floated de-SPACs do: ran from around $10 to $42 before settling back. Craig-Hallum initiated coverage with a Buy, framing $BRUN as a cheaper way to play neocloud demand than the larger names. The momentum is real; so is the balance-sheet question underneath it.

What is a neocloud, and what does Boost Run do?

A "neocloud" is a company that buys GPUs at scale and rents them out as cloud capacity, undercutting the hyperscalers on price for AI training and inference. Think of it as a landlord for accelerators: it raises capital, signs multi-year contracts with customers who need compute, buys the hardware, and earns the spread between the rent and the financing cost.

Boost Run delivers exactly this stack: GPU clusters, CPU nodes, managed Kubernetes orchestration, shared storage, networking fabric, and interconnects into the larger clouds, all through a console and API. Its stated differentiator is compliance: SOC 2 Type II, HIPAA, and ISO 27001/27701 at the operator level, aimed at regulated enterprise AI workloads that the bigger, faster-growing neoclouds tend to underserve. It was founded in 2023 and is based in Northbrook, Illinois.

How it makes money, and the financing gap

The revenue model is contracted compute: customers commit to multi-year terms, Boost Run provisions the GPUs, and revenue accrues over the life of the contract. The backlog of $940M (roughly three-year average terms) is the forward book of that revenue, and the gross margin on the services line is high (around 85%).

The problem is one line above the income statement: capital. To serve the backlog, Boost Run signed a $1.44B GPU hardware procurement agreement with Dell. At year-end 2025 it had roughly $9.7M of cash, a $21M working-capital deficit, and $39.7M of lease liabilities, and it took in about $112M from the SPAC trust on closing. A billion-plus of hardware does not get funded from that. The capital has to come from somewhere: equity raises (dilution), debt (interest), or vendor/lease financing (also debt, by another name).

This is the structural tension every neocloud lives with, but it is sharpest here because of scale. CoreWeave ($CRWV) runs the same model with around $20B of debt secured against a roughly $99B backlog. Boost Run has neither the balance sheet nor the contract base to borrow on those terms yet, which makes it the highest-beta, highest-risk way to own the trade.

Where it sits: the neocloud cohort

Boost Run lives in QA's AI-Capex Hyperscalers bubble as the closest existing home for a name that rents AI capacity rather than building its own models, and in the compute-capacity theme alongside its peers. The cohort is a clean comparison set: $CRWV (the largest, $2.08B in quarterly revenue), $NBIS (Nebius), $IREN, and Applied Digital ($APLD). Boost Run is the small-cap of the group, and it is priced at a discount to CRWV and NBIS on backlog, which is the bull's entire setup and the bear's "for a reason."

The useful way to place it: this is not a chip company and not a model company. It is the capacity layer, the part of the AI stack that converts capital and contracts into rentable GPU hours. Its fortunes track AI infrastructure demand and the cost of money, not silicon design.

The numbers

| Metric | Value | As of | | --- | --- | --- | | Market cap | ~$2.21B | 2026-06-04 | | Shares outstanding | ~31.9M (micro-float) | 2026-06-04 | | Contracted backlog | $940M (~3-yr terms) | 2026 | | Hardware commitment | $1.44B with Dell | 2026 | | 2026 revenue growth (fcst) | ~250% | 2026 guide | | Gross margin | ~85% | 2026 | | Cash | ~$9.7M (+ ~$112M trust) | 2025-12-31 | | Working-capital deficit | $21M | 2025-12-31 |

The growth rate is eye-catching and the backlog is real, but the cash line and the share count are the two numbers that decide the outcome. A ~31.9M-share float means equity raises move the stock hard; a $9.7M cash base against a $1.44B bill means raises are likely. Read the next few financing announcements more carefully than the next revenue print.

The bull case

• A $940M backlog and the $1.44B Dell agreement lock in both demand and hardware access at once, rare for a company this young.
• Inference-compute demand is the tailwind, and Craig-Hallum's Buy initiation frames BRUN as a discounted entry versus CRWV and NBIS.
• Roughly 250% forecast revenue growth for 2026 off a small base, with ~85% gross margin on the services line.
• The compliance posture (SOC 2 / HIPAA / ISO) targets regulated AI workloads the larger neoclouds chase less aggressively.

The bear case

• The financing gap is the whole risk: ~$9.7M of cash against a $1.44B commitment forces dilution or high-cost debt, and the micro-float makes dilution especially punishing.
• De-SPAC mechanics: warrant overhang, lockup expirations, and the kind of $10-to-$42 volatility that has nothing to do with the business.
• Valuation sits near 82x trailing sales with deeply negative operating and net margins, pricing flawless execution on three weeks of public history.
• Backlog customer concentration is not disclosed; one contract slipping would move the model.
• Sector risk: hyperscaler in-sourcing and custom silicon pressure the entire neocloud complex if AI capacity demand cools.

How to access

Boost Run trades on the Nasdaq as BRUN, a direct US listing. Unlike the mega-cap names, it is too new and too small to sit in the big index or semiconductor ETFs, so there is no fund wrapper that gives you incidental exposure: owning BRUN means owning the single stock, with all the de-SPAC volatility that implies. To trade it from a US-retail account alongside the rest of the AI-infrastructure names, see /stack/ibkr.

Rule-based alerts on $BRUN, the kind that fire on a curated level break or a bubble-correlation shift, are part of /pro.

What to watch

• The first financing moves: any equity raise, convertible, or debt facility tells you how the $1.44B gets funded and how much dilution comes with it.
• Backlog-to-revenue conversion: whether the $940M starts showing up as recognized revenue on the forecast path.
• Lockup expirations and warrant activity on the micro-float.
• The next quarterly print (the first as a public company) for the cash-burn and capex trajectory.
• A bubble-level shift: if the hyperscalers bubble and the broader compute-capacity cohort derate together, BRUN moves with the group regardless of its own backlog.

Live data on this ticker: /stocks/brun. Price, ETF holdings, bubble correlation, bot positions.

Bubble context: /bubbles/hyperscalers. The cluster this name belongs to and how it's moving.

QuantAbundancia is educational research. Nothing here is investment advice. See /disclosures.