Go further · Chapter 11 · 11a · Real-world case
How reliable are quantum computing roadmaps?
The short answer (tap to open)
Quantum companies publish roadmaps with dated milestones. Some deliver on time, some late, and some quietly move their targets. Comparing what each promised with what it delivered is the best guide to how much to trust its next promise. Watch for a shift from counting physical qubits to counting logical qubits: that's the field growing up.
What do the records show?
IBM has published roadmaps since 2020. It delivered the record-size chips it promised, 433 qubits (Osprey, 2022) and 1,121 qubits (Condor, 2023), then deliberately changed course towards fewer, better-connected qubits and error correction. Its current roadmap aims for Starling, 200 logical qubits, in 2029.
Quantinuum promised Helios for 2025 and delivered it in November 2025. IonQ reached its "#AQ 64" target three months early, in September 2025.
Rigetti promised a 100-plus-qubit system at 99.5% accuracy by the end of 2025, and delivered 108 qubits at about 99.1% in April 2026. Pasqal moved its 10,000-qubit target from 2026 to 2029.
Google set out six milestones in 2021. It has passed two, the second an error-correction prototype in 2023; the third, a long-lived logical qubit with one error in a million operations, is still ahead.
- IBM: A 433-qubit chip (Osprey). Promised 2022. On time: November 2022.
- IBM: A 1,121-qubit chip (Condor). Promised 2023. Not yet due, or not yet delivered.
- Quantinuum: Helios. Promised 2025. Not yet due, or not yet delivered.
- IonQ: #AQ 64. Promised about December 2025. Not yet due, or not yet delivered.
- Rigetti: 100-plus qubits at 99.5% accuracy. Promised the end of 2025. Not yet due, or not yet delivered.
- Pasqal: 10,000 qubits. Promised 2026. Not yet due, or not yet delivered.
Promise and outcome, in a table
| Promise | Promised | Outcome |
|---|---|---|
| IBM: A 433-qubit chip (Osprey) | 2022 | On time: November 2022. |
| IBM: A 1,121-qubit chip (Condor) | 2023 | On time: December 2023. |
| Quantinuum: Helios | 2025 | On time: November 2025. |
| IonQ: #AQ 64 | about December 2025 | Early: September 2025, three months early. |
| Rigetti: 100-plus qubits at 99.5% accuracy | the end of 2025 | Late: April 2026, 108 qubits at about 99.1%. |
| Pasqal: 10,000 qubits | 2026 | Moved to the end of 2029. |
What's changed in how roadmaps read?
Early roadmaps raced to bigger physical qubit counts. Recent ones promise logical qubits and error rates. That's a healthy sign: it's what chapter 8 says matters.
Check
Two quick ones
Collect
Your aha card
Promises are cheap; track records aren't.
Check what each company delivered last time.
Next optional page · 11b
Why doesn't qubit count tell you how good a quantum computer is?
Next optional pageCompany information is for education only. It isn't investment advice. Figures carry dates.