metaopt/torchopt
TorchOpt is an efficient library for differentiable optimization built upon PyTorch.
TorchOpt helps machine learning engineers and researchers build and train complex neural networks more efficiently. It takes network parameters and loss functions, and outputs optimized parameters, speeding up training for advanced models like those in meta-learning or reinforcement learning. This is designed for practitioners who develop and fine-tune machine learning models.
625 stars. Available on PyPI.
Use this if you are a machine learning practitioner who needs to implement advanced optimization techniques, especially for bi-level optimization problems or when experimenting with functional programming styles in PyTorch.
Not ideal if you are looking for a simple, off-the-shelf deep learning library and are not comfortable with advanced optimization concepts or functional programming paradigms.
Stars
625
Forks
42
Language
Python
License
Apache-2.0
Category
Last pushed
Mar 02, 2026
Commits (30d)
0
Dependencies
5
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/metaopt/torchopt"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Related frameworks
nschaetti/EchoTorch
A Python toolkit for Reservoir Computing and Echo State Network experimentation based on...
opthub-org/pytorch-bsf
PyTorch implementation of Bezier simplex fitting
gpauloski/kfac-pytorch
Distributed K-FAC preconditioner for PyTorch
pytorch/xla
Enabling PyTorch on XLA Devices (e.g. Google TPU)
stanford-centaur/PyPantograph
A Machine-to-Machine Interaction System for Lean 4.