cuStabilizer: A High-Performance Library for Stabilizer Quantum Simulators

Welcome to the cuStabilizer library documentation!

NVIDIA cuStabilizer is a high-performance library for stabilizer quantum simulators. cuStabilizer is a component of the NVIDIA cuQuantum SDK, which can be installed using instructions in Getting Started.

Functionalities of cuStabilizer are described in Overview with a usage guide provided under Examples.

cuStabilizer provides two sets of API:

1. Python API focused on ease-of-use.

2. C/C++ API focused on performance.

Key Features

• Pauli Frame simulation of noisy Clifford quantum circuits.

• Direct access to underlying GPU data structures for advanced integration.

• Circuit description that is easy to integrate with existing simulators.

Support

• Supported GPU Architectures: Ampere, Ada, Hopper, Blackwell

• Supported OS: Linux

• Supported CPU Architectures: x86_64, ARM64

Prerequisites

One of the following CUDA Toolkits and a compatible driver are required:

CUDA Toolkit	Minimum Required Linux Driver Version
CUDA® 12.x	>= 525.60.13
CUDA® 13.x	>= 580.65.06

Please refer to CUDA Toolkit Release Notes for the details.

Contents

• Release Notes
  • cuStabilizer v0.1.0
• Overview
  • Introduction to Pauli frame simulation
    • Trajectory-based noise simulation
    • Pauli frame simulation
    • Main data structures
  • Useful tips
    • Environment variables for logging
  • Citing cuQuantum
• Examples
  • Library initialization
  • Simple circuit simulation
• API Reference
  • cuStabilizer data types
    • Opaque Handle Types
    • Status and Enumerations
    • Data Structures and Integer Types
  • cuStabilizer functions
    • Library Initialization and Management API
    • Circuit API
    • Simulator API