torchquantum and quantum
These are competitors offering similar core functionality—both are hybrid quantum-classical ML frameworks—but targeting different deep learning backends (PyTorch vs TensorFlow), so practitioners typically choose one based on their existing ML infrastructure rather than using them together.
About torchquantum
mit-han-lab/torchquantum
A PyTorch-based framework for Quantum Classical Simulation, Quantum Machine Learning, Quantum Neural Networks, Parameterized Quantum Circuits with support for easy deployments on real quantum computers.
Supports statevector and pulse-level GPU simulation scaling to 30+ qubits, with dynamic computation graphs enabling interactive debugging. Integrates seamlessly with PyTorch's autograd for automatic gradient computation and batch tensorized processing, plus Qiskit for hardware deployment. Distinguishes itself through trainable parameterized gates, hybrid classical-quantum model construction, and measurement strategies supporting both analytical and stochastic sampling.
About quantum
tensorflow/quantum
An open-source Python framework for hybrid quantum-classical machine learning.
Integrates Cirq for quantum circuit design and qsim for high-performance simulation, while implementing quantum operations as native C++ TensorFlow Ops for seamless integration in the compute graph. Provides automatic differentiation of quantum circuits through multiple gradient methods (parameter shift, adjoint) and leverages Keras for defining quantum machine learning models, enabling researchers to scale quantum algorithm exploration across millions of circuit simulations.
Related comparisons
Scores updated daily from GitHub, PyPI, and npm data. How scores work