TurboDiffusion: 100–200× Acceleration for Video Diffusion Models

小彭老师领衔编写，现代C++的中文百科全书

Experiments with calculating Pi using different parallel frameworks

Original reference implementation of "3D Gaussian Splatting for Real-Time Radiance Field Rendering"

A unified library of SOTA model optimization techniques like quantization, pruning, distillation, speculative decoding, etc. It compresses deep learning models for downstream deployment frameworks …

Denoising Diffusion Probabilistic Models

Code for loralib, an implementation of "LoRA: Low-Rank Adaptation of Large Language Models"

Unofficial implementation of "Image Inpainting for Irregular Holes Using Partial Convolutions". Try at: www.fixmyphoto.ai

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.

This is an official repository of End-to-end Lane Shape Prediction with Transformers.

Simple ptq/qat workflow, based on https://github.com/NVIDIA/TensorRT/blob/release/10.2/quickstart/quantization_tutorial/qat-ptq-workflow.ipynb

Machine Learning Containers for NVIDIA Jetson and JetPack-L4T

High-efficiency floating-point neural network inference operators for mobile, server, and Web

On-device AI across mobile, embedded and edge for PyTorch

A batched offline inference oriented version of segment-anything

PyTorch native quantization and sparsity for training and inference

PyTorch Quantization Aware Training Example

CUDA implementation of Image Completion Using Global Optimization(Nikos Komodakis and Georgios Tziritas)

Official PyTorch implementation of the paper “LCCNet: Lidar and Camera Self-Calibration usingCost Volume Network”.

Re-Implementation of "Image Inpainting for Irregular Holes using Partial Convolution"

A tiny (803KB) but mighty timer in pure C. Supports clock, countdown, stopwatch, Pomodoro, and fully customizable tray animations (GIFs, CPU/Mem%).

CVPR 2025, EchoMatch: Partial-to-Partial Shape Matching via Correspondence Reflection

distributed builds for C, C++ and Objective C

Mirror only. Official repository is at https://git.zx2c4.com/wireguard-go

使用 rust 实现的飞连客户端

AutoMQ is a diskless Kafka® on S3. 10x Cost-Effective. No Cross-AZ Traffic Cost. Autoscale in seconds. Single-digit ms latency. Multi-AZ Availability.

FlatBuffers: Memory Efficient Serialization Library

LightGlue: Local Feature Matching at Light Speed (ICCV 2023)