The fundamental package for scientific computing with Python.

TensorLy: Tensor Learning in Python.

Constrained optimization toolkit for PyTorch

PyTorch Code for Energy-Based Transformers paper -- generalizable reasoning and scalable learning

A PyTorch library entirely dedicated to neural differential equations, implicit models and related numerical methods

High-Performance Symbolic Regression in Python and Julia

Lightweight coding agent that runs in your terminal

Library for Jacobian descent with PyTorch. It enables the optimization of neural networks with multiple losses (e.g. multi-task learning).

Mastering Diverse Domains through World Models

TextGrad: Automatic ''Differentiation'' via Text -- using large language models to backpropagate textual gradients. Published in Nature.

Model interpretability and understanding for PyTorch

Differentiable ODE solvers with full GPU support and O(1)-memory backpropagation.

Official implementation of DeepLabCut: Markerless pose estimation of user-defined features with deep learning for all animals incl. humans

Unofficial implementation of Titans, SOTA memory for transformers, in Pytorch

High accuracy RAG for answering questions from scientific documents with citations

System 2 Reasoning Link Collection

Learnable latent embeddings for joint behavioral and neural analysis - Official implementation of CEBRA

[ICLR2025] Halton Scheduler for Masked Generative Image Transformer

Vector (and Scalar) Quantization, in Pytorch

ImageBind One Embedding Space to Bind Them All

The repository provides code for running inference with the SegmentAnything Model (SAM), links for downloading the trained model checkpoints, and example notebooks that show how to use the model.

Official Implementation of Semantic Image Synthesis via Diffusion Models

[NeurIPS2022] Mind Reader: Reconstructing complex images from brain activities

Code base for MinD-Vis

Simple image captioning model

High-Resolution Image Synthesis with Latent Diffusion Models

Implementation of Vision Transformer, a simple way to achieve SOTA in vision classification with only a single transformer encoder, in Pytorch