#198198th of 242 in ML frameworks and librariesFor science, health and research, science

neuraloperator

PyTorch library of neural operator models, including Fourier Neural Operators

Stars
4k
License
MIT
Last commit
Oct 2026
Last release
Oct 2025
Language
Python
Stars, last 29 weeks+433
25 Apr 2026: 3,563 stars2 May 2026: 3,580 stars9 May 2026: 3,597 stars16 May 2026: 3,614 stars23 May 2026: 3,627 stars30 May 2026: 3,644 stars6 Jun 2026: 3,661 stars13 Jun 2026: 3,675 stars20 Jun 2026: 3,690 stars27 Jun 2026: 3,702 stars4 Jul 2026: 3,716 stars11 Jul 2026: 3,729 stars18 Jul 2026: 3,744 stars25 Jul 2026: 3,757 stars1 Aug 2026: 3,776 stars8 Aug 2026: 3,788 stars15 Aug 2026: 3,803 stars22 Aug 2026: 3,812 stars29 Aug 2026: 3,838 stars5 Sept 2026: 3,862 stars12 Sept 2026: 3,874 stars19 Sept 2026: 3,892 stars26 Sept 2026: 3,910 stars3 Oct 2026: 3,926 stars10 Oct 2026: 3,943 stars17 Oct 2026: 3,959 stars24 Oct 2026: 3,972 stars31 Oct 2026: 3,980 stars7 Nov 2026: 3,989 stars14 Nov 2026: 3,996 stars

3,563 on 25 Apr 20263,996 on 14 Nov 2026

From GitHub's daily star counts.

Checked on GitHub.

Overview

NeuralOperator is a PyTorch library for training neural operators, which learn mappings between function spaces and can be applied at different data resolutions. It includes the official Fourier Neural Operator (FNO) implementation, a Tucker-factorized variant (TFNO) and other operator architectures. Optional Weights & Biases logging is set up through a key file in the config directory.

Who it is for: Researchers and engineers training neural operators on PDE and scientific data

Strengths

  • Official FNO implementation, with a JMLR paper and a library paper behind it
  • TFNO uses Tucker factorization to cut Fourier layer parameters to about 10%
  • Installable from PyPI with pip install neuraloperator
  • Part of the PyTorch Ecosystem; MIT licensed

Weaknesses

  • Scoped to operator learning for PDE-style data, not general deep learning
  • README quickstart only constructs models; no training example is shown
  • Last tagged release is October 2025, so main may be ahead of PyPI

What it needs

  • Needs PyTorch, Weights & Biases (optional)
  • Models: FNO, TFNO

What changed

From the release notes. No flag means the note does not say so, not that the upgrade is safe.

Also in ML frameworks and libraries

See all 242
Also in ML frameworks and libraries
RankProjectScore
1transformersModel-definition library for pretrained text, vision, audio and multimodal models169.3k starsApache-2.094 out of 100
2ultralyticsPython library and CLI for training and running YOLO vision models63.5k starsAGPL-3.094 out of 100
3pytorch-image-modelsPyTorch library of image models, pretrained weights and training scripts37.3k starsApache-2.094 out of 100
#4pytorchPython tensor library with GPU support and tape-based autograd105.8k starscustom license93 out of 100
#5supervisionPython toolkit for loading, annotating and counting computer vision detections Live demo ↗ (opens in a new tab)51.7k starsMIT92 out of 100