Test Oratomic's 10,000-qubit claim before pricing it at $5.5B
A three-point test for judging Oratomic's $5.5B quantum valuation before proof arrives.
Before you decide whether to underwrite Oratomic's Series B, check whether the 10,000-qubit claim can be verified outside the company.
A March 31, 2026 Caltech-Oratomic publication argued that a utility-scale quantum computer, a machine large enough to be useful for real work, could operate with as few as 10,000 qubits, compared with the roughly 1 million qubits previously assumed necessary.
Oratomic disclosed a $475 million Series B on October 8, 2026, at a $5.5 billion valuation, lifting cumulative funding to $775 million and moving the company's value from $1.5 billion to that valuation.
Founded in 2025 and publicly launched on March 31, 2026, the company added a $300 million July raise to the October round, producing the total within roughly nine months of its debut.
ARCH Venture Partners, Spark Capital, Khosla Ventures, Index Ventures, General Catalyst, and Bezos Expeditions co-led the round.
Oratomic co-founder and CEO Dolev Bluvstein expects a working quantum computer by 2030.
The valuation is a bet on that smaller machine. A high valuation before a working machine reflects a lower risk of the core technical claim. The faster the funding, the more the price depends on the technical case being right. The work is to separate the company's story from evidence that can be checked.
Independent reproduction is the first gate
Reproduction is the direct way to test whether the smaller-machine case survives outside the original team.
Investors should look for work from groups with no commercial stake in the company. A useful check is whether another lab can reproduce the key result without relying on the original team's data.
The evidence gets stronger when the reproduction is peer-reviewed, independently analyzed, and not limited to a single demo. Press releases can announce a result, but they cannot replace a method another group can follow. A successful run is a start, not a verdict.
Ask whether the result changes the engineering problem, not just the headline. The strongest evidence would come from a group that did not help build the original claim. It should show the same performance under conditions that a skeptical engineer would call realistic. The result should survive a second reading by someone who did not write the paper.
Look for data that can be inspected, methods that can be repeated, and results that hold across runs. A reproduction that works only in a narrow setup leaves the technical threshold fragile. One that works in another setting makes the claim harder to dismiss.
What matters is whether the physics can be checked, not whether the company is promising a lot.
Rival targets and a held schedule are the remaining gates
Competitors moving their public targets toward the smaller-machine case would show the technical claim has gained traction.
Without that movement, the claim remains a company-specific bet. Rival revision matters because it changes the industry's map. A public roadmap, a paper, or a product plan counts more than a vague comment. Watch for changes in planning assumptions, not just commentary.
Teams still planning around much larger machines leave Oratomic's case isolated. Teams moving toward the smaller-machine case give the market a shared reference point.
The 2030 working-computer date remains a promise, not a proof. A held 2030 schedule requires milestones that keep the date credible.
A date slip should prompt a question about what changed: the hardware, the team, or the definition of a working machine. Schedule discipline is about credibility, not optimism.
Use the three-point threshold test before pricing the round
Signals that appear give the valuation a technical floor. A failed signal calls for a discount.
- Independent reproduction: another lab, with no commercial stake, reproduces the 10,000-qubit utility-scale result in peer-reviewed, inspectable work.
- Rival target revision: competitors move public roadmaps, papers, or product plans toward the smaller machine.
- Held 2030 schedule: Oratomic keeps the 2030 working-computer date credible with milestones, not just promises.