{Rethinking Transformer Connectivity: TLinFormer, A Path to Exact, Full Context-Aware Linear Attention

TLinFormer to TConstFormer: The Leap to Constant-Time Transformer Attention

Parallel Prefix Sum (Scan) with CUDA

Some common CUDA kernel implementations (Not the fastest).

cuTranspose: a library to transpose 3D arrays in Nvidia CUDA GPUs

A search and recommender system based on Elasticsearch, Neo4j, Flask, Apache

C++ extensions in PyTorch

RWKV, in easy to read code

High-Resolution Image Synthesis with Latent Diffusion Models

Finetune your VAE on private datasets!

Tiny AutoEncoder for Stable Diffusion (and other image models)

MixTeX multimodal LaTeX, ZhEn, and, Table OCR. It performs efficient CPU-based inference in a local offline on Windows.

Implementation of rectified flow and some of its followup research / improvements in Pytorch

AutoEncoder trained on ImageNet

Conan - The open-source C and C++ package manager

yolov5 + csl_label.(Oriented Object Detection)（Rotation Detection）（Rotated BBox）基于yolov5的旋转目标检测

Simple renderer for use with MuJoCo (>=2.1.2) Python Bindings.

3D mesh render without pain.

Octo is a transformer-based robot policy trained on a diverse mix of 800k robot trajectories.

ROS Wrapper for RealSense™ Cameras

Robotics Toolbox for Python

Powerful, mature open-source cross-platform game engine for Python and C++, developed by Disney and CMU

Open-source simulator for autonomous driving research.

Tools for converting ROS messages to and from numpy arrays

Windows 12 网页版，在线体验 点击下面的链接在线体验

(Flattened) Docker image for ros-humble-desktop-full

A little guide to help you install & manage NVIDIA GPU driver on your Ubuntu system