Pillar A · State as of 2026-10-08

Did any quantum startup come out of YC’s last three batches, and what does it promise?

As of October 8, 2026, the answer is zero. Across Y Combinator’s three most recent fully launched batches — Winter 2026 (214 companies, Extruct tracker), Spring 2026 (194, StartGround) and Summer 2026 (233, RoundFunded) — roughly 641 companies, not one is a quantum computing startup. YC’s own directory tags 3 quantum computing companies in its entire history of 5,000+ funded startups: Rigetti Computing (S14), Menten AI (W20) and Conductor Quantum (S24). The nearest miss inside the window is Dipole Labs (Summer 2026), a photonics company whose own site targets «AI and quantum infrastructure»; the nearest bet outside it is Sygaldry (Spring 2025 batch), which raised $139M in April 2026 (Quantum Computing Report) on claims that remain at promise level. Rosetta Quantum prints the zero — and every row behind it — with a source and a date.

→ Leer en español

State as of: 2026-10-08

Status as of: October 8, 2026. Batch counts come from the public trackers named on each row, read today; YC’s batch-filtered directory pages were not reachable from this run, so the directory evidence below is limited to the YC pages that were.

How many quantum startups came out of YC’s last three batches?

Zero out of roughly 641. The window here is the three most recent fully launched Y Combinator batches as of today: Winter 2026, with 214 companies in the Extruct batch tracker (updated September 2, 2026); Spring 2026, with 194 companies per StartGround (June 9, 2026); and Summer 2026, with 233 companies per RoundFunded (updated October 7, 2026). The Fall 2026 batch is in session now and its company list will publish around its demo day, so it will not be counted until it exists in public. Across those ≈641 tracked entries — 641 is our arithmetic on the three counts above — none describes itself as a quantum computing company, and none appears under YC’s own directory tag for the industry. That tag, read today, is its own data point: «Quantum Computing» lists exactly 3 companies out of the 5,000+ YC says it has funded since 2005 — Rigetti Computing (S14), Menten AI (W20) and Conductor Quantum (S24). The most recent of the three is from 2024.

Who is the nearest miss, and what does it actually say?

Dipole Labs, in the Summer 2026 batch, and the distinction matters. On its YC profile (read today) the one-liner is «AI-controlled optical switching for AI clusters», and the launch post says the company «builds high-speed optical circuit switches for AI data centers» — a classical photonics product. On its own site (read today), the framing widens: «We are building the photonic hardware that powers the next generation of AI and quantum infrastructure». So quantum appears as a stated target market for enabling hardware, not as the product: Dipole does not claim to build a quantum computer, and no quantum device or benchmark appears in its public record read today. The founders’ physics pedigree is real — the YC launch post says the CEO «built scalable quantum-optical hardware at ETH Zurich» — which makes the company a good marker of something this scan keeps finding: in 2026, quantum-trained founders entering YC are building classical photonics for AI, with quantum named as a market that will come later. In future tense, correctly.

What happened to YC’s actual quantum bets?

Three companies in 21 years, and their trajectory is the honest version of the industry’s. Rigetti Computing went through YC in Summer 2014 and raised a $2.5M seed that YC’s own blog announced at the time; it is now a public company (NASDAQ: RGTI) whose machines any reader can price per shot on AWS Braket — the catalog is in our price post from yesterday. Menten AI (W20) applies, per its YC directory entry, «quantum computing and machine learning to design protein drugs and enzymes» — no quantum-advantage claim from that work has entered our claims register. Conductor Quantum (S24) builds, per its one-liner, AI software that operates and tunes quantum computers — a bet on the software layer, not on qubits. And adjacent to the window sits the largest quantum bet in YC’s orbit: Sygaldry Technologies, in the Spring 2025 batch, founded by Chad Rigetti — the same founder YC backed in 2014 — announced $139M on April 14, 2026 (a $105M Series A led by Breakthrough Energy Ventures plus a $34M seed led by Initialized Capital, with Y Combinator listed among the round’s participants, per Quantum Computing Report).

What does Sygaldry claim, as written? Its YC profile (read today) says «quantum can provide exponential speedups for specific AI tasks» and that its servers will «combine multiple qubit types within a single, fault-tolerant architecture». Two reference points from the measured record, cross-linked without re-litigating: the public scoreboard of quantum speedups on generative-AI workloads stood at 0 as of our October 1 verdict (our generative-AI post), and the public record’s largest fault-tolerant computation ran on one logical qubit (our logical-qubit post). The $139M announcement cites no benchmark, no delivered hardware and no dated milestone (Quantum Computing Report, April 14, 2026). On the evidence scale Rosetta Quantum applies to every claim, a stated capability with no public artifact sits at the bottom rung: a promise — which is not an accusation, it is a date-stamped reading of what exists in public today. The company’s own wording is forward-looking, and this ledger will re-read it when an artifact appears.

Company	Batch	Claim as written (source, read 2026-10-08)	Public evidence state	Source
Rigetti Computing	S14	«full-stack hybrid classical/quantum computers» (YC directory)	shipped QPUs, publicly priced per shot; NASDAQ: RGTI	YC directory · YC blog 2014
Menten AI	W20	«quantum computing and machine learning to design protein drugs and enzymes» (YC directory)	no quantum-advantage claim in our register	YC directory
Conductor Quantum	S24	«AI that operates quantum computers» (YC directory)	software layer on top of QPUs; no advantage claim	YC directory
Sygaldry	Spring 2025	«exponential speedups for specific AI tasks»; «multiple qubit types within a single, fault-tolerant architecture» (YC profile)	$139M raised 2026-04-14; no benchmark, hardware or dated milestone cited	YC profile · QCR
Dipole Labs	S26	site: «photonic hardware that powers the next generation of AI and quantum infrastructure»; YC one-liner: «AI-controlled optical switching for AI clusters»	classical optical switching product; quantum is a stated market, not the product	YC profile · dipolelabs.com
Rest of W26 + Spring ’26 + S26	2026	—	0 quantum companies across ≈641 tracked entries	Extruct · StartGround · RoundFunded

Does YC even ask for quantum startups?

Not in the most recent request list we could verify. YC publishes «Requests for Startups» — the problem areas it explicitly invites founders to work on — and the Summer 2026 edition contained 15 ideas, from counter-swarm defense to inference chips for agent workflows. Quantum computing is not among the 15 (The VC Corner’s write-up of the S26 RFS, April 28, 2026; YC’s own RFS page was not reachable from this run). The demand side and the supply side of the batch match. A structural reading, declared as interpretation rather than measurement: quantum hardware does not fit accelerator economics. The capital that entered quantum in 2025 — $12.6B, 6.3× the 2024 figure, with ≈90% going to computing (McKinsey QTM 2026, via our usage map) — flows in nine-figure rounds to companies like PsiQuantum ($1B Series E, our photonics post) and Sygaldry ($139M before any public benchmark), not in batch-scale checks; and the software layer where a batch-sized quantum company could live captured only ≈6% of 2025 funding (QED-C, via our capital map). An accelerator batch is a sample of what founders believe they can build to revenue in months. In 2026, with the verified-advantage counter still at 0 (our usage map, October 6), quantum is not on that list — and that absence is information.

What would flip this zero?

A company in a future batch whose own one-liner claims quantum computing as its product. The Fall 2026 batch will publish its company list around its demo day, and this scan will re-run against it; the result will be published under this same URL with a new date, whichever way it falls. Two things this zero does not say, stated plainly: it does not say quantum computing is stalling everywhere — 2025 set the investment record cited above — and it does not say no quantum startups are being founded; it says the world’s most watched startup accelerator, the one that funded Rigetti in 2014, has launched ≈641 companies in its last three batches and zero of them are quantum computing companies. A zero without a specified test licenses very little — the rule Rosetta Quantum applied to its own detector in our calibration post — so this one comes with its test attached: the window, the counts, the trackers and the date are all printed above, and anyone can re-run the scan.

What we don’t know

We don’t know what is inside the 133 S26 entries that the RoundFunded tracker lists by name only, without descriptions — a quantum company could in principle hide there, and the zero would then be a one. We don’t know what YC declined: application data is private, so this scan counts funded companies, not founder supply. We could not read YC’s own batch-filtered directory or its RFS page from this run — the batch counts come from the three third-party trackers named above, which lag launches and can differ by a few entries, and the RFS evidence carries a newsletter’s byline, not YC’s. We don’t know whether Sygaldry has artifacts behind its claims that are simply not public — the reading above is of the public record only, and it will be re-read when the record changes. And we don’t know the Fall 2026 batch: it will publish when it publishes, and the scan will run again.