Stars
An efficient implementation of the NSA (Native Sparse Attention) kernel
Implementation of GeoDiff: a Geometric Diffusion Model for Molecular Conformation Generation (ICLR 2022).
Official Pytorch repository for "Diffusion Posterior Sampling for Linear Inverse Problem Solving: A Filtering Perspective", where FPS (Filtering Posterior Sampling) as well as its extension FPS-SMC…
[CVPR 2025 Oral] Improving Diffusion Inverse Problem Solving with Decoupled Noise Annealing
Domain-specific language designed to streamline the development of high-performance GPU/CPU/Accelerators kernels
Incremental engine for long horizon agents 🌟 Star if you like it!
🚀 Efficient implementations for emerging model architectures
EDM2 and Autoguidance -- Official PyTorch implementation
Python Fire is a library for automatically generating command line interfaces (CLIs) from absolutely any Python object.
Convert scikit-learn models to PyTorch modules
🐍 Geometric Computer Vision Library for Spatial AI
Codebase for evaluation of deep generative models as presented in Exposing flaws of generative model evaluation metrics and their unfair treatment of diffusion models
Vector (and Scalar) Quantization, in Pytorch
Implementation of Recurrent Interface Network (RIN), for highly efficient generation of images and video without cascading networks, in Pytorch
Repository of the paper "Accelerating Transformer Inference for Translation via Parallel Decoding"
Aim 💫 — An easy-to-use & supercharged open-source experiment tracker.
What would you do with 1000 H100s...
A guidance language for controlling large language models.
A playbook for systematically maximizing the performance of deep learning models.
[ICML 2023] Reflected Diffusion Models (https://arxiv.org/abs/2304.04740)
Official repo for consistency models.
Train to 94% on CIFAR-10 in <6.3 seconds on a single A100. Or ~95.79% in ~110 seconds (or less!)
Fixes macOS Preview garbled annotations
PyTorch extensions for high performance and large scale training.