Skip to content
Quantum.world

The Signal

Infleqtion reports 30 entangled logical qubits on Sqale

Infleqtion says its Sqale neutral-atom processor entangled 30 logical qubits in one computation using 80 physical qubits. Its earnings presentation also reports circuit-level fidelity and output yields for error detection and loss correction.

Infleqtion reports 30 entangled logical qubits on Sqale

One computation, 80 physical qubits

Infleqtion says its Sqale neutral-atom quantum processor entangled 30 logical qubits in a single computation, using 80 physical qubits. The company reported the milestone in its third-quarter results on November 16, describing it as the 2026 target achieved.

The result concerns a computation on hardware, rather than a simulated circuit. In a technical account, Infleqtion said the experiments were completed in August using its Superstaq software to compile circuits for Sqale.

A circuit spanning all 30

The company’s account describes an instantaneous quantum polynomial-time, or IQP, circuit with entangling gates connecting all 30 logical qubits. It involved approximately 1,000 physical quantum operations. The company says its experiment used 80 atoms grouped into ten blocks of eight; its technical write-up describes code preparation and a distance-2 encoding used in the subsequent experiments.

To assess the result, Infleqtion compared observed measurement outcomes with those expected from the circuit. Its blog reports a hit fraction of approximately 25%, roughly 1,000 times the uniform-random baseline. The company says there were more than a billion possible 30-bit outcomes, while the ideal circuit produces 262,144 of them, about 0.024% of the possibilities.

Gate savings and reported yields

Infleqtion credits an AI-assisted discovery with reducing the physical two-qubit gates required for a key logical entangling operation from eight to four. The company says the resulting double-CZ operation enabled the 30-logical-qubit state with 136 physical two-qubit gates.

The November earnings presentation gives additional performance figures for the circuit. From 88,000 shots, it lists 21% circuit-level fidelity and 39 shots for error detection; for correction for loss, it lists 7% fidelity and 1,408 shots, marked as 36 times the yield. The presentation reports these as separate rows, rather than as a single fidelity or output figure for the whole computation.

Next roadmap target

Infleqtion says it is working toward longer computations and a target of 100 logical qubits in 2028. The reported milestone is a 30-logical-qubit entangled computation on Sqale; the company’s reported fidelity and yield figures describe the specific circuit and error-handling methods it presented.

Sources

  1. Infleqtion, Inc. (SEC filing)
  2. Infleqtion
  3. Infleqtion

Companies in this story

Related stories

Explore

More articles