IBM's Quantum Roadmap: From Nighthawk to Starling in 2029
The plan
IBM publishes one of the most detailed public roadmaps in the field. As reported:
- Nighthawk: a 120 qubit processor introduced in November 2025.
- Loon (2025), Kookaburra (2026), Cockatoo (2027): test processors stepping toward modular, error corrected designs. Kookaburra is described as the first modular processor that stores and processes encoded information.
- Starling (2029): 200 logical qubits running circuits of 100 million gates, built in Poughkeepsie, New York.
- Blue Jay (2033): a later, larger system on the roadmap.
The technical bet
IBM favors qLDPC error correcting codes, which it says can reduce the number of physical qubits needed per logical qubit compared with the surface code. See the surface code and error correction.
Verifiable advantage by end of 2026
IBM has publicly said it aims to demonstrate verifiable quantum advantage by the end of 2026. Whether it does so is one of the biggest stories to watch in the next few months. Read how advantage claims work.
Skeptic's view
Roadmaps are targets. Analysts still question the timeline and near term business value. Check IBM's own roadmap page for the current status.
Where to watch
The IBM Quantum Developer Conference runs November 11 to 13, 2026 in Chicago, per IBM. See the events calendar.
What has already shipped
At its November 2025 announcements, IBM introduced Nighthawk, a 120 qubit processor with 218 tunable couplers. IBM said it expected later versions to handle circuits of around 7,500 gates in 2026 and up to 10,000 gates in 2027, as reported by technology press. IBM also reported that its decoding system could correct errors in real time, within 480 nanoseconds, which matters because error correction only works if the classical decoder keeps up. Loon was described as an experimental processor that tests the components needed for fault tolerance, including many more layers of wiring. See real time error correction and AI for decoders and calibration.
How IBM plans to check its own claim
IBM said it is working with outside groups, including Algorithmiq, the Flatiron Institute and BlueQubit, on an open community tracker for quantum advantage candidates, so claims can be tested by classical researchers. This is a healthy sign. Earlier IBM results were challenged by classical methods, as covered in advantage vs utility. A claim that survives a public challenge is worth far more than one that does not.
Status as of 2026-10-09
- Nighthawk and Loon: announced in 2025, per IBM and press coverage.
- Kookaburra (2026): still listed as a 2026 target. We could not find a report in the sources we read confirming its delivery, so treat it as unconfirmed and check IBM's roadmap page.
- Verified advantage by end of 2026: still a stated goal. We could not confirm a community accepted case as of today. The IBM developer conference in November 2026 is a likely place for updates. See what to expect at IBM's conference.
- Starling (2029): still the target, in Poughkeepsie, New York.
What Starling would mean
Two hundred logical qubits running a hundred million operations would be enough to try problems beyond classical reach in chemistry and materials, although it would not be enough to break modern encryption by itself. IBM describes the design as modular, with error correcting memory connected to processing units, which is also why the intermediate Kookaburra and Cockatoo chips matter. See resource estimation and logical qubits.
How IBM compares
Google has emphasized measured error correction scaling (Willow), while IBM has emphasized a detailed public schedule and a software ecosystem. See superconducting makers compared and the innovators guide. Many people find IBM's openness about dates useful, because it makes the plan testable. This is education, not investment advice, and there is no connection to the QNT memecoin.
Sources and further reading
- IBM quantum roadmap
- TMCnet: IBM lays out roadmap for 2029
- IBM: Quantum Developer Conference 2026
- PostQuantum.com: IBM Loon and Nighthawk
- Nextgov: IBM unveils updated quantum computing products
Reported as of 2026-10-09. Company roadmaps are targets and often slip. Check each company's own announcements.
Frequently asked questions
What is IBM Starling?
A planned fault tolerant quantum computer with 200 logical qubits and 100 million gates, targeted for 2029.
When is IBM's quantum developer conference?
November 11 to 13, 2026 in Chicago, according to IBM.
What is qLDPC?
A family of error correcting codes IBM uses to reduce the qubit overhead of error correction.
Is the roadmap guaranteed?
No. These are company targets and can change.
What is Nighthawk?
A 120 qubit IBM processor introduced in November 2025 with 218 tunable couplers, aimed at running more complex circuits.
Has IBM delivered Kookaburra?
We could not confirm it as of 2026-10-09. It is listed as a 2026 target. Check IBM's roadmap page.
How will IBM prove quantum advantage?
By submitting candidates to an open community tracker with partners, so classical researchers can try to match them.
Keep reading
- The Biggest Innovators in Quantum Computing in 2026
A tour of the companies pushing quantum forward in 2026: IBM, Google, Quantinuum Ltd, IonQ, Microsoft, PsiQuantum, Rigetti, D-Wave, Pasqal, Xanadu, IQM and NVIDIA. - Quantum Catalysts to Watch 2026 to 2030: When the Story Gets Loud
The milestones, deadlines and events that could crank up the quantum narrative through 2030, from IBM's 2026 advantage target to the 2030 cryptography cutoffs. - Quantum Error Correction Explained
Qubits are fragile, so quantum computers need error correction. Learn how logical qubits are built and why this is the key challenge. - Superconducting Qubits Explained in Depth
How do superconducting qubits work? A clear look at circuits, microwave control, strengths, weaknesses and open questions in widely used quantum hardware.
All Quantum industry, people and AI guides | Back to top | Search the site
Main pages: Quantum computing explained | Quantum and crypto | Companies | Quantum news | Glossary