Keep track of advancements in every frontier
No dispatches yet.
Aquila and gate-based systems
Photo: QuEra Computing, Inc. · CC BY 4.0 (opens commons.wikimedia.org)

Aquila and gate-based systems

by QuEra ComputingUS

PilotStage 3 of 5

Aquila is on AWS Braket. Gate-based systems delivered to research centers.

Updated 9 Sep 2025Checked 25 Sep0 updates this week

Milestones

No announced next step
  1. Aquila available on Amazon BraketNov 2022Complete.
  2. 48 logical qubits (with Harvard)6 Dec 2023Complete.
  3. Continuous operation of a 3,000-qubit array (Harvard/MIT)Sep 2025Complete.
  4. System delivered to AIST in Japan2025Complete.

Most important updates

  • 9 Sep 2025
  • Feb 2025
  • 30 Apr 2024
  • 7 Feb 2024
  • 6 Dec 2023

Current obstacles

  • Slow cycle timesMoving and imaging atoms takes milliseconds, so error-correction cycles are much slower than on superconducting chips.

Physics limits

  • Moving atoms takes timeShuttling atoms between zones takes hundreds of microseconds per move, and moving faster heats them. Computations with many rearrangements run far slower than superconducting ones.
  • Analog mode can't be error-correctedAquila runs continuous physics directly, with no discrete steps where errors could be checked. Noise accumulates over the run, and no known method corrects it in analog mode.
  • Readout is slow and can lose atomsAtoms are read by imaging their fluorescence, which takes milliseconds and can push atoms out of their traps, slowing each error-correction round.

How it works

3 parts
Aquila's optical table: the lasers and optics that cool the atoms, hold them in tweezers and drive them into Rydberg states
Aquila's optical table: the lasers and optics that cool the atoms, hold them in tweezers and drive them into Rydberg statesPhoto: QuEra Computing, Inc. · CC BY 4.0 (opens commons.wikimedia.org)
Analog

Aquila: programmable physics

Up to 256 rubidium atoms are placed in any 2D pattern and driven so their interactions mimic a magnetic material or an optimisation problem.

Shuttling

Moving atoms mid-program

In digital mode, laser tweezers carry atoms between storage, gate and readout zones, so any qubit can be paired with any other.

Parallel

One pulse, many gates

A single laser pulse applies the same gate to hundreds of atom pairs at once, which suits error-correcting codes with repeated structure.

Update log

5 updates

Tue 9 Sep 2025

  • Minor: PressFunding

Feb 2025

  • Major: PressFunding

Tue 30 Apr 2024

  • Major: PressCommercial

Wed 7 Feb 2024

  • Minor: PressCommercial

Wed 6 Dec 2023

  • Major: PaperResearch

About QuEra Computing

The team behind Aquila and gate-based systems

QuEra Computing

Neutral-atom computers, Aquila, logical qubits

QuEra, founded in 2018 by Harvard and MIT physicists, builds neutral-atom computers from rubidium atoms in movable laser-tweezer grids. With Harvard it ran 48 logical qubits in 2023. Its Aquila machine is on AWS.

  • Raised over $230M in February 2025 with Google and SoftBank Vision Fund 2. NVentures invested later in 2025.
  • Harvard-led team with QuEra ran algorithms on 48 logical qubits (Nature, Dec 2023).
  • Delivered a system to Japan's AIST.
Founded
20188 yrs
Headquarters
United States
Status
Private
Valuation
private
Raised
$230M1 round
Last round
Venture round · $230MFeb 2025
Works in
QuantumNeutral atom
Coverage
1 program · 5 updateslatest 9 Sep 2025checked 25 Sep
Lead investors
GoogleSoftBank Vision Fund 2
Primer