Qᴜᴀʟᴛʀᴀɴ

Quantum computing hardware continues to advance. While not yet ready to run quantum algorithms with thousands of logical qubits and millions of operations, researchers have increased focus on detailed resource estimates of potential algorithms—they are no longer content to sweep constant factors under the big-O rug. These detailed compilations are worked out manually in a tedious and error-prone manner.

This is the documentation for Qᴜᴀʟᴛʀᴀɴ (quantum algorithms translator): a set of abstractions for representing quantum programs and a library of quantum algorithms expressed in that language.

Note: Qualtran is in beta. Some algorithms or library functionality may be incomplete or contain inaccuracies. Open issues or contact the authors with bug reports or feedback, and subscribe to qualtran-announce@googlegroups.com to receive the latest news and updates!

Fundamentals

Qualtran has tools and abstractions for expressing and reasoning about quantum algorithms, programs, and subroutines. Our hosted language consists of Python objects representing operations (Bloq), quantum data types, and algorithms. We provide protocols for simulating algorithms, estimating resource requirements, drawing diagrams, and more.

The Fundamentals section provides documentation for these tools and abstractions.

Bloqs Library

Qualtran also contains a library of quantum subroutines as well as high-level quantum programs. These primitives can be imported from the qualtran.bloqs subpackage and are documented in the Bloqs Library docs section.

API Reference

Each class, method, and function in the library is documented in the API Reference.

Installation

Basics

Qualtran is a Python package. If you already have a working Python environment,

pip install qualtran

will install the library and its dependencies. Qualtran uses Graphviz to draw diagrams, which can’t be installed using the Python package manager pip. Follow the Graphviz installation instructions to set it up. Qualtran requires Python 3.10 or newer.

Installing from source

Source code is hosted on GitHub. You may wish to inspect the source code for the bloqs provided in the library or interactively execute the Jupyter notebooks included in the repository. In this case, you can clone the git repository and install from source.

git clone https://github.com/quantumlib/Qualtran.git
cd Qualtran/
pip install -e .

If you plan on developing Qualtran or you would just like to install all optional and developer dependencies, you can run

pip install -r dev_tools/requirements/envs/dev.env.txt

from the repository root.

Setting up a Python environment

There are many ways to install Python and Qualtran’s dependencies. This section will provide a walkthrough of one particular way.

First, download and install Miniconda. This is a small tool that lets you install and manage Python environments. It supports different versions of Python in each environment, and can be installed without Administrator privileges. After following the installation instructions for Miniconda on your platform, you can create and activate an environment for Qualtran:

conda create -n my-qualtran-env python=3.11
conda activate my-qualtran-env

That code snippet will create a new environment with the given name (choose whatever you like) and install Python into it. Each time you start a new terminal you must first activate your environment

conda activate my-qualtran-env

You can use conda directly to install Graphviz:

conda install graphviz

At this point, you have an isolated, modern Python installation with Graphviz installed. From here, you can follow the basic installation instructions.

pip install qualtran

Next Steps

You should be able to import qualtran from a Python command prompt.

import qualtran

If this is successful, you can move on to learning how to write bloqs or investigate the bloqs library.