D-Wave Quantum
The Only Dual-Platform Quantum Bet

Executive Summary

D-Wave is the only public quantum company running two platforms at once: a commercial quantum-annealing business that is live and selling today (the 4,400-qubit Advantage2 system, sold as access through the Leap cloud and as on-prem systems), and a gate-model program brought in-house in January 2026 through the $550M acquisition of Quantum Circuits, Inc. (dual-rail qubits), with an initial system targeted for 2026. The annealing franchise is genuinely differentiated, purpose-built for optimization and sampling, and in March 2025 D-Wave published a Science paper claiming "quantum supremacy" on a magnetic-materials simulation versus the Frontier supercomputer² (a claim that immediately drew classical-method rebuttals). The catch is valuation: FY2025 revenue was just $24.6M (at an 82.6% gross margin) against a ~$8.5B market cap, a trailing EV/Sales multiple north of 322×: and revenue is lumpy because system sales dominate the print. Our base-case discounted EV/Sales model implies ~$15 vs the $22.98 close (the Street is far more bullish, 15 analysts at "Strong Buy", ~$37 average target), a clean dual-platform-execution bull case supports ~$40, a niche-annealing bear case ~$7.

The Four Pillars of the Debate

Investment Thesis

The QBTS debate pits a genuinely differentiated commercial-quantum franchise against a valuation that already prices in near-flawless multi-year execution. D-Wave has to convert annealing momentum (Leap usage, record bookings) into durable, less-lumpy revenue and land a credible gate-model system, or the ~322× trailing multiple compresses toward a specialized-vendor band. We frame three scenarios using a discounted EV/Sales model, QBTS is pre-profit, so a DCF is inappropriate.

Targets are the output of a simplified discounted EV/Sales model with subjective assumptions, illustrative, not analyst price targets. See sections 08 and 10 to adjust. Probability weights are 30/40/30 to reflect the high outcome dispersion of a pre-profit deep-tech name with both a steep growth path and a stretched starting valuation. The base sits below the current price by design, the disciplined model does not support today's tape as a base case.

Business Overview

D-Wave is the pioneer and only commercial vendor of quantum annealing: a special-purpose architecture that finds low-energy solutions to optimization and sampling problems by evolving a system of qubits toward its ground state. It is not a general-purpose (gate-model) computer; it is purpose-built for a class of problems, combinatorial optimization, materials/spin-glass simulation, sampling, where it is already commercially useful today. The flagship is Advantage2: 4,400+ qubits, 20-way connectivity (up from 15), ~40% higher energy scale, ~4× lower noise and roughly double the coherence of its predecessor.

What makes D-Wave unique among public quantum names is that it is now dual-platform. Alongside annealing, it is building a gate-model (general-purpose) system, having closed its $550M acquisition of Quantum Circuits, Inc. (the Schoelkopf-founded dual-rail-qubit pioneer) on January 20, 2026: $300M in stock plus $250M cash, and targets bringing an initial gate-model system to market in 2026 (a 17-physical-qubit system), scaling on a published roadmap to ~175 physical qubits by 2028 and ~100 logical qubits by 2032. The gate-model approach centers on high-fidelity, error-detecting dual-rail qubits (an order of magnitude fewer physical qubits per logical qubit), local on-chip cryogenic control, and multi-chip superconducting packaging. Annealing is the revenue today; the gate-model is the optionality the market is increasingly paying for.

Two Architectures, One Company: Annealing vs Gate-Model

The dual-platform structure is D-Wave's central differentiator: annealing is a real, near-uncontested commercial business for optimization, while the gate-model program, if it delivers, gives D-Wave a seat at the general-purpose table the rest of the field is racing toward. The risk is that annealing stays a niche and the gate-model arrives late or behind better-funded incumbents.

Industry & Quantum Cycle

Quantum computing is really two races. The general-purpose (gate-model) race, IONQ, IBM, Google, Quantinuum, AWS, is in the NISQ era (Noisy Intermediate-Scale Quantum): 100–2,000 physical qubits, limited error correction, very few production workloads, with fault tolerance (logical qubits running arbitrarily long algorithms) as the gating milestone, Google's Willow (2024) was the first below-threshold error-correction demo; IBM targets fault tolerance ~2029. The special-purpose (annealing) race is, in commercial terms, a category of one: D-Wave's annealer is already useful today for optimization and sampling, without waiting for fault tolerance.

D-Wave is the only public name straddling both. Its annealing business gives it real revenue now, while its 2026 gate-model program (built on the now-owned Quantum Circuits dual-rail technology, acquired Jan 2026) is its bid to also compete in the general-purpose race. The bull case is that owning both is uniquely valuable; the bear case is that annealing is a niche and the gate-model is a late, underfunded entrant against IBM and Google.

The Competitive Set

In annealing, D-Wave's real competition is classical optimization and the contested supremacy benchmark, not other quantum vendors. In the gate-model race it is the most concentrated threat, IBM and Google can outspend it 100-to-1 with explicit fault-tolerance roadmaps. Owning both platforms is the differentiator; executing both is the challenge.

Where QBTS Sits in the Cycle

The Opportunity (TAM)

Quantum computing attacks problems classical machines cannot solve in tractable time, drug & materials discovery, optimization, logistics, finance, cryptography. The market is small today but inflecting: independent forecasters project the quantum-computing provider market growing from roughly $3.5B in 2025 to ~$20B by 2030 (~42% CAGR), with McKinsey estimating up to ~$72B of annual QC revenue by 2035 and BCG projecting $450–850B of cumulative economic value by 2040.

D-Wave's angle on that TAM is unusual. Its commercial annealing business addresses the optimization and sampling slice, a real, monetizable subset available today, but not the whole prize, which is dominated by general-purpose (gate-model) computing. The bull case is that (a) annealing's optimization slice is larger and stickier than skeptics assume, and (b) the 2026 gate-model program lets D-Wave also compete for general-purpose dollars. Put together, that underwrites ~2–7% of the 2030 provider TAM: ~$1.45B of 2030 revenue in the bull case, ~$850M in the base, ~$480M in the bear. The bear view is that annealing stays niche and the gate-model lands late, capping share toward the low end.

Financial Health

FY2025 GAAP revenue was $24.6M, up +179% from $8.8M in FY2024, at an 82.6% gross margin (up from 63%), but the growth came from system sales, so it is real and lumpy at once. The full-year adjusted EBITDA loss was $71.8M; the GAAP net loss was $355M, but roughly $270M of that is non-cash warrant/derivative remeasurement that scales with the share price (the stock ran to ~$46 in 2025), so the cleaner read of the operating burn is the adjusted net loss of $84.5M. Q1 CY2026 made the lumpiness point: revenue fell to $2.9M (−81% YoY) because the prior-year quarter had carried a $12.6M annealing-system sale and this one did not. The GAAP net loss was $18.4M in Q1 and the adjusted EBITDA loss was $32.8M: both stepping up as operating expense rises with the gate-model build. Read the run-rate, not any single revenue quarter.

The balance sheet is the strongest part of the story. $588M cash & equivalents at Q1 CY26 (10-Q dated 2026-03-31), up 93% YoY after 2025 ATM raises, with effectively no traditional debt. Against a ~$130M annualized cash operating loss, that implies ~4 years of runway absent a raise: enough to fund the gate-model launch and several years of commercial scaling. The base case still bakes in +20% net dilution from ATM programs across the 2026→2030 window; the bear case +25%.

Versus IONQ, the closest public comp, D-Wave carries a smaller revenue base but a more commercially-diversified one (135 customers, a recurring Leap cloud line), and a comparably well-capitalized balance sheet. The dilution math in section 08 is material to the per-share fair-value calculation, but the cash position converts dilution from a survival question into a timing-and-discipline question.

Cash Runway: Multi-Year, Funds the Gate-Model Build

Bookings & Backlog

Bookings are where the D-Wave inflection is most visible, and the cleanest counter to the lumpy revenue line. Q1 CY2026 bookings were a record $33.4M, up roughly 2,000% YoY and more than the company booked in all of FY2025 ($18.7M). Backlog is finally building: remaining performance obligations (RPO) of $42.4M at Q1 CY26 (+563% YoY), with ~54% expected to convert within 12 months and ~71% within two years. Two post-year-end deals anchor the print: a $20M system purchase by Florida Atlantic University and a $10M, two-year QCaaS enterprise license with a Fortune-100 company.

Unlike the gate-model peers, D-Wave's bookings are commercially anchored rather than government-grant anchored, it recognized revenue from 135 customers in 2025, including 76 commercial enterprises and 28 Forbes Global 2000 names, with Advantage2 access usage up 314% YoY. The watch item is conversion: bookings and RPO are leading indicators, but revenue recognition is still lumpy and system-sale weighted, so a strong bookings quarter does not guarantee a strong revenue quarter (Q1 26 is the cautionary example). The bull case needs the recurring Leap line to compound so revenue smooths out behind the bookings momentum.

Identifiable Contracts & Partners

Bookings and RPO are the leading indicators that the commercial thesis is working; revenue recognition lags and stays lumpy. The line to watch quarter to quarter is recurring Leap / QCaaS subscription growth, the variable that would let revenue smooth out behind the bookings momentum.

Valuation & Comps

QBTS trades at roughly 322× trailing (FY25) sales (and ~640× on the depressed trailing-twelve-month figure, since Q1 26's $2.9M replaced Q1 25's $15.0M). The multiple is eye-watering primarily because the denominator is so small, D-Wave's enterprise value of ~$7.91B is a large-but-not-absurd number for a deep-tech name the market believes in; it is divided by $24.6M of revenue, which makes the headline ratio look broken.

The honest reframing is the forward-2030 multiple. At the base-case ~$850M 2030 revenue and a 12× exit EV/Sales, the implied 2030 enterprise value is ~$10.2B, discounting it back four years at 14% with +20% net dilution lands on ~$15/share, vs the current $22.98 close. The bull case ($1,450M at 17×, +10% dilution, 12% WACC) takes that to ~$40; the bear ($480M at 10×, +25% dilution, 18% WACC) to ~$7. That is the EV/Sales-compression chart on the summary page in plain language, and it is why our disciplined base sits below the tape.

Versus IONQ, the closest direct comp, QBTS's trailing multiple (~322×) is richer because the revenue base is smaller, but the gap narrowed after FY2025's +179% step-up. On forward 2030E base, both compress materially; the key difference is that D-Wave's compression rests partly on a gate-model program that has only just shipped its first system, exactly the risk the bear case prices. Note the divergence from the sell side: 15 analysts rate QBTS "Strong Buy" with a ~$37 average / $40 median target (range $30–$45), they underwrite the bull ramp; our disciplined base ($15) deliberately does not.

Valuation Model: Discounted EV/Sales

Pick a ~2030 revenue, an exit EV/Sales multiple, a discount rate and net dilution; the model discounts the implied future enterprise value back four years, adds net cash, and divides by diluted shares. The dilution slider matters, D-Wave funds operations through ATM programs, but the $588M Q1 CY26 cash position means the base case can be funded with only +20% net dilution across the 2026→2030 window. Use the presets, then move the sliders.

Revenue Ramp & Implied EV ($M)

Sensitivity: Implied Price ($)

Catalysts & Roadmap

Roadmap Milestones

Scenario Targets

This calculator uses the same discounted EV/Sales math at a fixed 14% WACC and 4-year horizon. Move the inputs to see the implied price and a rough scenario-likelihood read. The dilution range spans 0–80% across the model window, D-Wave's ATM history makes the dilution path a major swing factor even with the $588M cash buffer. Note that even aggressive revenue assumptions struggle to reach today's ~$23 price, the disciplined-base-below-market story in one tool.

Bull vs Bear

Technicals & Positioning

QBTS is among the most volatile equities in the quantum cohort. The 52-week range, $12.75 to $46.75: is roughly a 3.7× spread, anchored by the late-2025 quantum-sector rally (D-Wave's "supremacy" result, Google Willow) that carried into a 2025 momentum peak near $46.75, a pullback to a ~$12.75 low at end-March 2026, and a choppy recovery to the low-$20s by June 2026. The two June-2026 federal quantum executive orders sparked a brief rally that quickly reversed, QBTS gave back ~8% on June 24 as momentum traders (who had piled into call options, ~5:1 over puts) took profits. At $22.98 the stock sits ~51% below its 52-week high.

Short interest runs heavy, typically ~20%+ of the float: reflecting the stretched valuation and the structural-dilution thesis. That positioning cuts both ways: it amplifies downside on negative milestones and equity-raise announcements, but it sets up sharp short-squeeze rallies on positive surprises (a gate-model demo, a large system sale, a quantum-policy headline, or a sector-wide re-rate). The shares trade with high retail ownership and move tightly with IONQ and RGTI.

Risk / Reward Tool

Ask the Thesis AI-assisted checking…

Describe a scenario in natural language; the assistant returns a structured impact analysis against D-Wave's dual-platform roadmap (annealing + gate-model), the contested supremacy claim, lumpy-revenue and dilution math, and EV/Sales scenario framework. Powered by Claude via a Cloudflare Worker proxy (Anthropic key held server-side).

Note: The assistant reasons from the dashboard's data snapshot and thesis sections, it does not browse the web or access real-time fundamentals beyond what's in data.js. Treat its responses as scenario-modeling support, not as primary research.

Risks