Quantinuum reports 99.9995% logical fidelity on Helios trapped-ion system
Quantinuum says its Helios trapped-ion computer hit 99.9995% logical fidelity using a color code, 10x better than its September result and within Apollo's target range.
The story in focus
Quantinuum reported that its Helios trapped-ion computer achieved 99.9995% logical fidelity, or five errors per million operations, using a color code. The company said this is ten times lower than its own September 2026 record and within the target range for Apollo, its planned fault-tolerant machine expected by 2029.
- Helios reached 99.9995% logical fidelity, equal to five errors per million operations.
- The result is 10x lower than Quantinuum's September 2026 record and 100x lower than the best published result on other hardware modalities.
- Each increase in code distance reduced remaining logical errors by a factor of 12, the company said.
The result and its comparison
Quantinuum said its Helios system achieved 99.9995% logical fidelity, equivalent to five errors per million operations, using a well-known color code within its error-correction architecture. The company reported the error rate as ten times lower than its own record from September 2026 and 100 times lower than the best published result across other hardware modalities, which it attributed to Google Quantum AI on a superconducting system.
Quantinuum said its successive breakthroughs used different codes on the same hardware. On Helios, its Helix approach used 10 physical qubits per logical qubit, compared with up to 37 for the color code, while the color code delivered lower logical error rates.
Scaling evidence
The study showed logical errors falling exponentially as code distance increased. Quantinuum said each tested increase reduced remaining logical errors by a factor of 12, which it described as evidence of increasingly effective error suppression as the code grows.
The company said the accuracy is now in the target range for Apollo, its planned universal fault-tolerant quantum computer expected to be available by 2029. It also said the result suggests fault tolerance could be reached with significantly fewer qubits than other modalities require.
What remains ahead
Quantinuum described its architecture as combining tailored AI decoders, all-to-all qubit connectivity, high-fidelity physical operations and hardware-level error correction. Features include adaptive syndrome extraction and chain-mapping to connect codes suited to different tasks.
The company cautioned in forward-looking statements that the results were achieved in experimental settings and may not be replicated at larger scale, in logical computation or in commercial use, and that Apollo may not arrive on the expected timeline.
How much lower is the new logical error rate than Quantinuum's September 2026 result?
Quantinuum said the Helios logical error rate is ten times lower than its own September 2026 record, and 100 times lower than the best published result across other hardware modalities.
Sources & further reading
Help keep the record accurate. Send a correction.
