For years, quantum computing headlines focused on one easy number: how many qubits a machine contained. In 2024, that started to change. When I look at the latest quantum computing companies’ breakthroughs in 2024, the most important advances were not simply bigger processors.
Companies began proving that qubits could become more reliable, errors could be reduced through encoding, and different quantum architectures could move closer to commercially useful computation.
Google, Microsoft, Quantinuum, IBM, Atom Computing, IonQ, D-Wave, and Rigetti all reached important milestones during the year. Collectively, their work showed that the next stage of quantum competition will depend on qubit quality, logical error rates, scalability, connectivity, and the ability to execute increasingly complex circuits.
Why 2024 Became a Turning Point for Quantum Computing
Quantum computers are extremely sensitive to noise. A physical qubit can lose its quantum state or suffer an operational error long before a complicated calculation finishes.
That is why simply increasing physical qubit counts cannot solve the industry’s biggest problem.
A more promising approach is to combine multiple physical qubits into logical qubits that can detect or correct errors. At the same time, companies are increasing gate fidelity, improving chip architectures and developing processors capable of running deeper circuits.
Several breakthroughs in 2024 moved directly toward these goals.
Microsoft and Quantinuum Made Logical Qubits Dramatically More Reliable
One of the year’s most important announcements arrived in April 2024.
Microsoft applied its qubit-virtualization system to Quantinuum’s 32-qubit H2 trapped-ion processor. The collaboration created four logical qubits from 30 physical qubits.
More importantly, Microsoft and Quantinuum reported logical error rates approximately 800 times lower than corresponding physical error rates. The teams also ran more than 14,000 independent circuit experiments without an error.
The work demonstrated active syndrome extraction, which allows a quantum system to identify errors without destroying the encoded quantum information.
Progress continued later in the year. By September, Microsoft and Quantinuum had demonstrated 12 logical qubits using an upgraded 56-qubit H2 system.
This rapid improvement made logical-qubit performance one of the defining themes of quantum computing in 2024.
Google Willow Demonstrated Below-Threshold Error Correction

Google Quantum AI produced one of the most widely discussed breakthroughs of the year when it announced Willow in December 2024.
The superconducting processor demonstrated something quantum researchers had been pursuing for decades: error rates declined as the size of Google’s error-corrected qubit arrays increased.
Google tested progressively larger encoded arrays and reported that the error rate was reduced by roughly half each time the system scaled. This is known as operating “below threshold,” an essential requirement for scalable quantum error correction.
Willow also completed a random-circuit-sampling benchmark in under five minutes. Google estimated that performing the same benchmark on a leading classical supercomputer would require around 10 septillion years.
That benchmark does not mean Willow can suddenly replace conventional supercomputers for everyday business applications. Its greater importance lies in demonstrating that error-corrected superconducting systems can potentially improve as they scale instead of becoming progressively less reliable.
Microsoft and Atom Computing Reached 24 Entangled Logical Qubits
Neutral-atom computing also produced a major 2024 result.
In November, Microsoft and Atom Computing announced that they had created and entangled 24 logical qubits by applying Microsoft’s qubit-virtualization technology to Atom Computing’s neutral-atom hardware.
The companies also performed computation using 28 logical qubits created from 112 physical qubits. According to Microsoft, the logical implementation produced more accurate results than the corresponding physical-qubit computation.
The collaboration also demonstrated techniques for detecting and correcting atom loss, an important challenge in neutral-atom architectures.
This result expanded the logical-qubit race beyond trapped ions and superconducting circuits and demonstrated why neutral atoms have become one of the industry’s most closely watched hardware platforms.
IBM Improved Heron Rather Than Chasing Qubit Count Alone
IBM’s progress in 2024 illustrates the broader industry shift toward performance rather than headline qubit numbers.
IBM introduced a revised Heron processor with 156 qubits in July 2024. Heron R2 added improvements designed to increase coherence and stability while continuing to use tunable couplers that help suppress unwanted interactions between qubits.
At IBM’s 2024 Quantum Developer Conference, the company demonstrated quantum circuits involving approximately 5,000 two-qubit gates on its hardware and software stack.
IBM also reported continued improvements in gate errors and processing speed while working on modular coupling technologies intended to connect processors as future systems become larger.
The strategy matters because useful quantum computers will require much more than thousands of poorly controlled qubits. They will need processors capable of maintaining useful fidelity through increasingly deep computations.
IonQ Crossed the 99.9% Two-Qubit Fidelity Mark

IonQ’s trapped-ion roadmap also advanced significantly during 2024.
In September, the company reported greater than 99.9% two-qubit gate fidelity on a next-generation barium development platform. The experiment involved optimized gates on a two-ion chain and was intended to support IonQ’s future barium-based quantum systems.
Higher native fidelity is particularly important because better physical operations can reduce the error-correction overhead required for future fault-tolerant quantum computers.
IonQ also ended 2024 with its Forte Enterprise machine operating at an algorithmic qubit performance level of #AQ36 at its European facility in Switzerland.
Rather than focusing entirely on physical qubit numbers, IonQ’s approach highlights another competitive metric: how accurately a system can perform useful quantum operations.
D-Wave Advanced Its Advantage2 Quantum Annealing Platform
D-Wave occupies a different part of the quantum computing landscape because its commercial systems are primarily based on quantum annealing rather than universal gate-model computation.
In early 2024, D-Wave released a prototype Advantage2 processor containing more than 1,200 qubits.
According to the company, the system increased qubit connectivity from 15-way to 20-way, increased energy scale by more than 40%, doubled coherence time and demonstrated substantial performance improvements on selected optimization problems compared with its previous Advantage platform.
This distinction is important. D-Wave’s breakthrough should not be evaluated using exactly the same criteria as Google’s or IBM’s gate-model processors.
Its progress instead demonstrates how quantum annealing continues developing as a specialized technology for optimization, sampling and related workloads.
Rigetti Improved Fidelity With the 84-Qubit Ankaa-3
Rigetti closed 2024 with the launch of its 84-qubit Ankaa-3 superconducting quantum computer.
The new processor incorporated redesigned cryogenic hardware, changes to its qubit circuit layout and improved frequency targeting.
Rigetti reported approximately 99% median fidelity for its universal iSWAP two-qubit gates while also demonstrating 99.5% median fidelity using specialized fSim gates. The company said it had roughly halved its two-qubit error rate during 2024.
For Rigetti, these improvements strengthened its strategy of combining superconducting qubits with modular architectures that could eventually connect multiple chips into larger processors.
What These Quantum Breakthroughs Actually Changed
The most significant lesson from the latest quantum computing companies breakthroughs 2024 is that the industry began changing the way progress is measured. Raw qubit counts still matter, but they are increasingly only one part of the equation.
Researchers and businesses are now paying much closer attention to logical qubits, two-qubit gate fidelity, circuit depth, connectivity, error correction, processor modularity, and the computational quality delivered by an entire quantum system.
Much like becoming a pro streamer requires more than simply having high-end equipment, meaningful progress in quantum computing depends on how effectively all parts of the system work together.
That shift is healthy for the industry because a million unreliable physical qubits would be far less valuable than a smaller architecture capable of sustaining accurate logical operations.
Which Companies Appeared Strongest in 2024?

There was no single universal winner because the companies were pursuing different technologies.
Google produced one of the strongest demonstrations of below-threshold error correction on superconducting hardware.
Microsoft and Quantinuum delivered major progress in logical-qubit reliability.
Microsoft and Atom Computing expanded reliable logical computing into neutral-atom systems.
IBM focused on deeper circuits, improved Heron processors and modular scaling.
IonQ pushed trapped-ion fidelity beyond 99.9% on its barium development platform.
D-Wave improved the performance and connectivity of quantum annealing hardware.
Rigetti reduced errors while upgrading its superconducting processor architecture.
Together, these developments made 2024 less about determining which company had the largest processor and more about discovering which architectures could eventually produce useful, fault-tolerant quantum computation.
Frequently Asked Questions (FAQs)
1. What were the biggest quantum computing breakthroughs of 2024?
The largest breakthroughs included Google’s below-threshold Willow error correction, Microsoft and Quantinuum’s highly reliable logical qubits, Microsoft and Atom Computing’s 24 entangled logical qubits, IBM’s improved Heron processor, and significant fidelity improvements from companies such as IonQ and Rigetti.
2. Which quantum computing company made the biggest breakthrough in 2024?
There is no objective single winner. Google achieved an important superconducting error-correction milestone, while Microsoft and its hardware partners Quantinuum and Atom Computing produced some of the strongest logical-qubit demonstrations.
3. Why are logical qubits more important than physical qubit counts?
Logical qubits encode quantum information across multiple physical qubits so errors can be detected or corrected. Reliable logical qubits are necessary for running long, complicated quantum algorithms.
4. What do the latest quantum computing companies’ breakthroughs in 2024 mean for the future?
They suggest that competition is moving toward reliability and scalable error correction. The next major milestones will depend on increasing logical-qubit numbers while lowering logical error rates enough to run commercially and scientifically valuable algorithms.
The Bigger Picture
When I look back at 2024, I see a quantum industry beginning to mature. Bigger machines still attracted attention, but the most meaningful achievements came from making quantum information more reliable.
Google showed that error rates could decline as an encoded superconducting system grew. Microsoft, Quantinuum and Atom Computing demonstrated increasingly capable logical qubits. IBM pushed toward deeper circuits, while IonQ and Rigetti improved physical-qubit fidelity. D-Wave continued advancing a different but commercially relevant quantum annealing architecture.
The race is therefore no longer simply about who can manufacture the most qubits. Similar to optimising a network for gaming, where overall performance depends on latency, stability, and efficient resource management rather than raw speed alone, quantum computing progress depends on the reliability and coordination of the entire system.
The companies most likely to shape the next era of quantum computing will be those that can transform fragile physical qubits into dependable logical computation at a scale that solves problems classical computers cannot.