Neutral Atom

QuEra Computing

256-qubit neutral atom processor. Analog quantum simulation and a landmark 48-logical-qubit error correction demonstration.

About QuEra

Rubidium atoms in optical tweezers

QuEra Computing was founded in 2020 as a spinout from Harvard and MIT, building on years of research into Rydberg atom arrays. The company is headquartered in Boston and has established itself as the leader in neutral atom quantum computing, a technology that uses individual atoms as qubits rather than fabricated superconducting circuits.

QuEra's Aquila processor holds 256 Rubidium-87 atoms, each trapped by a tightly focused infrared laser beam called an optical tweezer. The atoms can be arranged in arbitrary 2D geometries, giving the hardware a reconfigurable connectivity that no superconducting device can match. When two atoms are brought close enough together, a strong interaction called the Rydberg blockade prevents both from being excited simultaneously. This blockade is the physical mechanism behind both the analog Hamiltonian simulation and the digital entangling gates.

In 2023, QuEra, Harvard, and MIT demonstrated 48 error-corrected logical qubits using the transversal gate approach, the largest demonstration of error-corrected quantum computing at the time. This result validated neutral atoms as a leading platform for fault-tolerant quantum computing and established QuEra as a key player in the path toward practical quantum advantage.

Hardware Specifications

System specs at a glance

Specification Value
Current systems Aquila (256 qubits, analog; 60 logical qubits demonstrated 2026); Gemini (gate-based, in development)
Qubit technology Rubidium-87 atoms in optical tweezers
Operating modes Analog (Aquila, public) and digital gate-based (Gemini, coming)
Qubit count (Aquila) 256 neutral atom qubits
Connectivity All-to-all within coherence range (reconfigurable)
Analog programming Rydberg Hamiltonian via drive and detuning pulses
Gate fidelity (digital target) >99% with local addressing (Gemini roadmap)
Cloud access Amazon Braket (Aquila), QuEra research partnerships
Primary SDK Bloqade (Python and Julia) for Aquila; standard SDKs for gate-based

Use Cases

Where QuEra hardware excels

Getting Started

Run your first analog simulation on Aquila

  1. Create an AWS account

    The easiest way to access Aquila is via Amazon Braket. Create a free AWS account at aws.amazon.com. New accounts receive AWS Free Tier credits. Braket simulators are free within the first hour per month; Aquila QPU time is billed per shot.

  2. Install Bloqade (recommended) or the Braket SDK

    pip install bloqade

    Bloqade is QuEra's native SDK for programming Aquila. It provides a high-level Python (and Julia) interface for specifying Rydberg Hamiltonian pulse sequences and atom geometries. You can also use the Amazon Braket SDK directly if you prefer. 

    pip install amazon-braket-sdk

  3. Define an atom geometry and pulse sequence with Bloqade

    from bloqade import start
from bloqade.atom_arrangement import Square

# 3x3 square lattice, 5 micrometers spacing
geometry = Square(3, lattice_const=5.0)

# Build a simple adiabatic sweep program
program = (
    start
    .add_positions(geometry)
    .rydberg.detuning.uniform
    .linear(start=-10, stop=10, duration=4.0)
    .amplitude.uniform
    .constant(value=15.0, duration=4.0)
)

    This defines a 9-atom array with a detuning sweep and constant drive, implementing a basic adiabatic protocol. The Rydberg blockade enforces constraints that map to optimization problems.

  4. Run on the local emulator first

    # Emulate locally before spending QPU shots
batch = program.bloqade.python().run(100)
report = batch.report()
print(report.counts())

    Bloqade includes a fast Python-based emulator. Always test and tune your program locally before submitting to Aquila. The emulator is free and runs without AWS credentials.

  5. Submit to Aquila via Braket

    # Submit to Aquila QPU via Amazon Braket
batch = program.braket.aquila().run_async(100)
report = batch.report()
print(report.counts())

    Set your AWS credentials via aws configure before submitting. Each shot costs approximately $0.01. Start with small shot counts (100-200) to control costs while validating your program.

  6. Apply for research access

    Academic researchers can apply for direct QPU access and collaboration through the QuEra Academic Research Program. This provides access to capabilities and systems not available via Braket, including early access to Gemini when it becomes available.

Pricing

Access costs

Aquila is billed per shot via Amazon Braket. There is no subscription or reservation fee for standard Braket access. Research program access is free for qualifying academics.

Learning Resources

Tutorials and reference docs