Home › Compare › IonQ vs Quantinuum
Head to headIonQ vs Quantinuum
The two leading trapped-ion companies. Quantinuum leads on published accuracy today; IonQ is betting on chips that control ions electronically and promises bigger machines sooner. IonQ's main headline number is its own benchmark.
Can they be compared? Yes for accuracy and qubit counts: both use trapped ions. Be careful with IonQ's "#AQ" (algorithmic qubits) figures: it's a benchmark IonQ defined and reports itself, so we don't rank on it unless QubitBenchmark has graded it.
Side by side
| Topic | IonQ | Quantinuum |
|---|---|---|
| Approach | Trapped ion | Trapped ion |
| Most accurate | #1 | |
| Error-corrected qubits | #1 | |
| Most qubits | #4 | #5 |
| Fastest | ||
| Use it today | #2 | |
| Latest machine | Tempo (described as 100 qubits); Superion 256 on order for 2027 | Helios (98 qubits, launched November 2025) |
| Can you use it? | Forte Enterprise on Amazon Braket and IonQ's cloud; Tempo installed for customers | Quantinuum's cloud, or installed on site |
| Headline accuracy claim | Over 99.99% two-qubit fidelity with electronic control (company-reported, 2025) | 99.921% two-qubit fidelity across all 98 qubits on Helios (company-reported, 2025) |
| Next big promise | 256 physical qubits: target was 2026, deliveries now planned for 2027 | Sol, up to 100 logical qubits, in 2027 (said to be on track in August 2026) |
| Delivery record | Hit its own #AQ targets early; broad quantum advantage by 2025 and 256 qubits in 2026 have both slipped | Helios delivered on schedule in 2025; Sol and Apollo (2029) reported on track |
How they differ
Both companies use trapped ions: single charged atoms held in place by electric fields, which make very accurate qubits. That makes this a fair fight on accuracy.
Quantinuum focuses on low error rates across a whole machine. Its Helios computer lets any qubit interact with any other, and its error-correction results are among the strongest reported. Its roadmap is short and specific: Sol in 2027, Apollo (fully fault-tolerant) in 2029.
IonQ is moving from laser-controlled ions to chips that control ions electronically, through its purchase of Oxford Ionics. It argues this can be mass-manufactured. Its roadmap is more ambitious, aiming for millions of qubits by 2030, but its 2026 target of 256 qubits has moved to 2027, and its "broad quantum advantage by 2025" goal wasn't met.
Who's ahead
On this month's rankings, see the table above. Beyond the numbers, Quantinuum's delivery record is the stronger of the two. IonQ's headline benchmark, #AQ, is its own measure, which is why we treat its results with care until QubitBenchmark grades them.
Bottom line: Quantinuum has the stronger delivery record and the best-documented accuracy across a whole machine. IonQ makes bolder promises and has bought its way into new technology, but its biggest targets have slipped.
Sources
- Quantinuum announces commercial launch of new Helios quantum computer, Quantinuum, 5 November 2025
- Quantinuum reports second quarter 2026 results, Quantinuum
- IonQ launches Superion product line, IonQ
- IonQ achieves 99.99% two-qubit gate performance, The Quantum Insider, 21 October 2025
- IonQ roadmap, IonQ