pennylane

Why we included this project

PennyLane treats quantum circuits as differentiable programs rather than as gate sequences you simply run. That design choice means you can train hybrid quantum-classical models with familiar autodiff tooling such as PyTorch, TensorFlow, or JAX, so teams exploring quantum machine learning and variational algorithms stay inside the ecosystem they already know. The library also covers quantum chemistry, with Hamiltonian simulation and resource-estimation utilities alongside the core circuit-building API, and its pluggable device layer lets one codebase target simulators or hardware from multiple vendors. A large collection of research demos and interactive tutorials makes it approachable for newcomers, which makes this a practical place to prototype whether near-term quantum methods can help with an optimization or ML problem before moving to real devices.