Neural Quantum States
NetKet provides state-of-the-art Neural-Network Quantum states, and advanced learning algorithms to study many-body quantum systems.
Jax-powered
Netket is based on Jax, therefore you can run on CPUs, GPUs and TPUs, using Sharding or MPI to distribute among multiple nodes.
Interoperable
NetKet can easily interoperate with other tools such as
OpenFermion or
QuTiP.
Multi-framework
NetKet’s built in models are implemented using
Flax, but it is also compatible with
Equinox,
Haiku and can be made compatible with future frameworks.
Easy to use
NetKet has an high-level syntax that is easy to use, while exposing low level objects that can be combined together to develop novel algorithms.
Open source
NetKet is a collaborative effort by a group of researchers. We are constantly looking for new contributors. Do get in touch with us through the
Github Discussions or through our Slack Channel
