Amazon Web Services’ Ocelot is a superconducting quantum-chip prototype built to test a fault-tolerant-computing architecture based on “cat qubits.” Its design aims to reduce the hardware resources needed for quantum error correction, but the published results do not establish Ocelot as a fault-tolerant, general-purpose quantum computer. AWS has not announced Ocelot as a purchasable product or a device available through Amazon Braket.
What is Amazon’s Ocelot quantum chip?
AWS announced Ocelot on February 27, 2025. The AWS Center for Quantum Computing, based at the California Institute of Technology, developed it as a first-generation prototype. Its purpose is to test a hardware design for quantum error correction, a key requirement for building more reliable quantum computers.
Ocelot is not a conventional processor intended to run consumer applications. It is an experimental device that combines quantum circuits and error-detection hardware to investigate how a future fault-tolerant system might be built.
How Ocelot’s cat-qubit design works
Cat qubits store information in oscillator states
Ocelot uses superconducting cat qubits. Rather than encoding information in the more familiar two-state form of a basic qubit, a cat qubit represents information in states of a microwave oscillator. The design intrinsically suppresses one important class of errors: bit flips, in which a qubit’s encoded value changes.
Outdated Drivers Are Slowing You Down
One free scan finds every outdated or missing driver and matches the right update for your exact hardware.Free scan · exact hardware matchPC Slower Than It Used to Be?
A free scan shows the junk files, broken settings and background clutter dragging Windows down - then fixes them in one click.Free scan · Windows 10 & 11#1 Best Overall
That protection does not eliminate every error. In particular, a system still needs to detect and manage other error types, including phase flips. Ocelot tests how to combine the cat qubits’ built-in protection with additional error-correction hardware.
Error correction is built into the chip architecture
The prototype combines five data cat qubits, five buffer circuits that stabilize them, and four additional qubits used for error detection. The components are distributed across two bonded silicon microchips. The design puts error correction into the hardware architecture from the outset rather than treating it solely as a layer added after the qubits are designed.
What the Ocelot experiment demonstrated
Amazon Science and AWS researchers reported bit-flip times approaching one second. They also tested error behavior on subsets of the prototype. In a repetition-code experiment, increasing the code distance from three to five reduced the measured logical phase-flip error rate.
The reported total logical error rates were 1.72% per correction cycle for distance three and 1.65% per cycle for distance five. These remain nonzero error rates measured per cycle, so the results are evidence about a prototype’s error-correction behavior—not proof that Ocelot can run arbitrary computations reliably at scale.
Free tools Windows power users keep installed
One-click scans. No signup required.
How Ocelot compares with surface-code approaches
AWS researchers reported that the distance-five Ocelot code used nine qubits, compared with 49 qubits for a comparable surface-code device. That is a comparison for the reported code and device, not a universal count for every cat-qubit or surface-code implementation.
| Comparison point | Ocelot result or design | What the comparison does—and does not—show |
|---|---|---|
| Qubit approach | Superconducting cat qubits, with information encoded in microwave-oscillator states | A hardware approach that intrinsically suppresses one important error class; it does not make all errors disappear. |
| Reported distance-five qubit count | Nine qubits for the reported Ocelot code | AWS researchers compared this with 49 qubits for a comparable surface-code device. The figures apply to that reported comparison, not all implementations. |
| Reported total logical error rate | 1.65% per correction cycle at distance five; 1.72% per cycle at distance three | These are prototype measurements, and the rates remained above zero. |
| Access to Ocelot hardware | AWS has not announced public cloud access to Ocelot | AWS names Amazon Braket as a way to explore quantum computing, but its announcement does not list Ocelot as an available device. |
Qubit count alone cannot establish which architecture is better. A fuller comparison would also consider logical error rates at matched code distances, how each system scales, fabrication demands, and whether hardware is publicly accessible. Ocelot’s reported results support evaluating its error-correction overhead; they do not establish that it outperforms surface-code systems across those measures.
Rank #4
What AWS’s cost and timeline figures mean
AWS said Ocelot could reduce the cost of implementing quantum error correction by up to 90% compared with current approaches. That is the company’s claim, not an independently established industry-wide result from a deployed computer.
Oskar Painter, AWS director of Quantum, said future chips built according to the Ocelot architecture could cost as little as one-fifth of current approaches. He also said AWS believes the approach could accelerate its timeline to a practical quantum computer by up to five years. Both figures are forward-looking estimates about future scaling, not measured savings or a demonstrated delivery schedule for Ocelot itself.
Recommended Free Tools
Can you buy or use Ocelot?
No Ocelot purchase option or public cloud-device listing is identified in AWS’s announcement. AWS directs scientists, developers, and students interested in quantum computing to Amazon Braket, a managed service that provides access to third-party quantum hardware, high-performance simulators, and software tools. That is a route to explore quantum computing, but it is not evidence that Braket provides access to Ocelot.
Ocelot’s status as of June 2026
In an AWS update dated June 15, 2026, the company said its Center for Quantum Computing continued developing superconducting cat-qubit devices such as Ocelot and described that work as complementary to other quantum modalities. The update did not give a production release date or a retail channel.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
