Wondering which types of qubits matter? Superconducting circuits are fast, trapped ions hold their state longest, and photons travel best. You’ll compare them best by error rates and coherence, not qubit count alone.
This guide sits within our wider coverage of quantum computing, and it focuses on the hardware.
| Item | Value |
|---|---|
| Superconducting operating temperature | Approximately 15 millikelvin |
| Superconducting coherence time | Typically microseconds |
| Superconducting raw gate speed | 1-100 MHz range |
| Trapped-ion two-qubit gate fidelity (IonQ prototype, October 2025) | 99.99% |
| Photonic: PsiQuantum fault-tolerant system target | By 2029 |
| Commercially available QPUs (as of August 2026) | 56 from 24 manufacturers |
These figures come from the MIT Quantum Index Report benchmarking chapter and The Quantum Insider’s overview of quantum computer types. They can change as new devices and results appear, so check the sources before you rely on them.
What makes a good qubit?
A useful qubit has to hold its state, be controlled and measured accurately, and interact with other qubits, according to The Quantum Insider. Each hardware approach balances those needs differently.
Four properties come up again and again when you compare qubits:
- Coherence: how long a qubit holds its quantum state before its environment erases it.
- Fidelity: how accurately gates, readout and state preparation work.
- Gate speed: how long a single-qubit or two-qubit operation takes.
- Connectivity and manufacturability: how qubits link up and how easily the hardware can be built.
Gate speed and coherence together set how many operations a system can complete before the qubit becomes ineffective, per the MIT Quantum Index Report. Two-qubit quantum gates deserve special attention, because the same report calls their fidelity among the most critical metrics, as they are usually the bottleneck in large circuits.
Superconducting qubits: fast gates, short coherence
Superconducting qubits are electrical circuits cooled to approximately 15 millikelvin, colder than deep space. IBM, Google and Rigetti all use this approach.
The circuits can be patterned onto silicon wafers using established chip processes, which gives the approach a potential manufacturing advantage as systems grow. The trade-off is a short coherence time, typically measured in microseconds. Adding more qubits also increases the engineering challenges around control electronics, wiring and heat management.
On speed, the MIT Quantum Index Report says superconducting machines operate at raw gate speeds in the 1-100 MHz range. Fast gates help offset the short coherence, but they do not remove the limit.
Two milestones show where the approach is heading, as reported by The Quantum Insider:
- In December 2024, Google’s Willow chip demonstrated below-threshold error correction on real hardware.
- IBM’s roadmap describes a Starling system designed to demonstrate 200 logical qubits and 100 million quantum gate operations.
Trapped-ion qubits: long coherence, hard to scale
Trapped-ion qubits start with individual atoms that have an electron removed, giving them a net electrical charge. Electromagnetic fields hold the ions in a vacuum chamber, and lasers manipulate their internal energy states. IonQ and Quantinuum both use this approach, according to The Quantum Insider.
Because the qubits are individual atoms of the same element, they are naturally identical. That helps trapped-ion systems reach high gate fidelity. Their gates typically run in microseconds rather than the nanoseconds common in superconducting systems, but the qubits keep coherence for much longer.
The MIT Quantum Index Report goes further: trapped-ion and, to a slightly lesser degree, neutral-atom systems show dephasing times several orders of magnitude longer than superconducting qubits. That is an inherent advantage for applications that need longer coherence.
Recent results show the payoff, as reported by The Quantum Insider:
- In October 2025, IonQ reported 99.99% two-qubit gate fidelity in a prototype.
- In March 2026, Quantinuum reported computations using 94 error-protected logical qubits, achieving beyond-break-even performance.
Scaling is the main challenge. Adding more ions to a single trap makes the system harder to control, and connecting several traps adds communication overhead between modules.
Photonic qubits: built for networks, hard to entangle
Photonic qubits encode information in properties of individual photons, such as polarization, path or arrival time. Optical parts like beam splitters and phase shifters route and transform the photon states to perform operations.
Photons interact weakly with their environment, which is both the main advantage and the central challenge. Weak interaction means photons resist decoherence and travel through standard optical fiber with low loss, so they suit quantum networking. It also makes photons difficult to entangle with each other, which complicates the two-qubit gates needed for general computation, per The Quantum Insider.
Many photonic components can run at room temperature, although single-photon detectors typically still need cooling.
The same source gives two company examples. In June 2025, Xanadu demonstrated on-chip generation of GKP states, an error-correctable way to encode photonic qubits, on a silicon nitride chip. PsiQuantum closed a $1 billion Series E in September 2025 and is targeting a fault-tolerant system by 2029 using silicon photonic chips made at GlobalFoundries (source).
How the three types compare
No single type wins on every measure. The table summarizes what each approach offers and where it struggles, based on the two sources above.
| Type | Qubit encoded in | Strength | Main challenge |
|---|---|---|---|
| Superconducting | States of current or charge in an electrical circuit | Fast gates; fabrication compatible with semiconductor processes | Short coherence; wiring and heat as qubit counts grow |
| Trapped-ion | Internal energy states of charged atoms | High gate fidelity; long coherence | Control gets harder with more ions per trap |
| Photonic | Polarization, path or arrival time of photons | Resists decoherence; suits fiber networking | Photons are difficult to entangle |
The speed-versus-coherence pattern is the clearest split. Systems with fast gates, such as superconducting qubits, generally have shorter coherence times than trapped ions or neutral atoms, which have slower gates but far longer coherence, according to the MIT Quantum Index Report.

Why qubit count alone misleads
A system with 1,000 noisy qubits may be less useful for some tasks than a system with 50 more reliable qubits, The Quantum Insider notes. Error rates tend to rise alongside qubit count.
IBM’s path shows the point. It followed its 1,121-qubit Condor in 2023 with the higher-performing 133-qubit Heron, then the 120-qubit Nighthawk in 2025, a device built for circuit quality rather than scale (MIT Quantum Index Report).
The same report describes hardware progress as fracturing into two strategies: chasing raw qubit scale, or prioritizing circuit quality and fidelity. As of its August 2026 data, it counted 56 commercially available processors from 24 manufacturers, with no clear frontrunner among the hardware types.
It also states that most experts agree quantum computers remain below classical computers for all useful tasks, as of that August 2026 data. For what it would take to change that, see our explainer on quantum advantage.
Frequently asked questions
Which type of qubit is best?
Show answer
None is, according to the MIT Quantum Index Report. Its August 2026 analysis found that superconducting, trapped-ion, neutral-atom and photonic systems each show distinct strengths and trade-offs without a clear frontrunner. The best choice depends on the task.
Why do superconducting qubits need such low temperatures?
Show answer
At around 15 millikelvin, certain materials conduct electricity with zero resistance and the circuit exhibits quantum behavior, according to The Quantum Insider. Most quantum designs also need heavy isolation, because any stray interaction destroys the quantum state.
Do photonic qubits need cooling?
Show answer
Many photonic components can operate at room temperature. Single-photon detectors typically still require cooling, per The Quantum Insider.
Are there other types of qubits?
Show answer
Yes. As of The Quantum Insider’s August 2026 overview, six hardware approaches are competing, including neutral-atom, topological and annealing systems. Topological quantum computing remains experimental and contested, and annealers target optimization problems rather than general-purpose algorithms (source).



