MNN is a blazing fast, lightweight deep learning framework, battle-tested by business-critical use cases in Alibaba. Full multimodal LLM Android App:[MNN-LLM-Android](./apps/Android/MnnLlmChat/README.md)

校招、秋招、春招、实习好项目！带你从零实现一个高性能的深度学习推理库，支持大模型 llama2 、Unet、Yolov5、Resnet等模型的推理。Implement a high-performance deep learning inference library step by step

Vulkan/CUDA/HIP/OpenCL/Level Zero/Metal Fast Fourier Transform library

Caffe for Sparse Convolutional Neural Network

Blazing-fast Expression Templates Library (ETL) with GPU support, in C++

🌠 Bloom Simulation Software for Windows

High Performance Computing (HPC) and Signal Processing Framework

Active Convolution

Reverb effect using hybrid impulse convolution

An example of C++ extension for PyTorch.

This project implements a convolution kernel based on vivado HLS on zcu104

This project is about convolution operator optimization on GPU, include GEMM based (Implicit GEMM) convolution.

Zero Config Convolver

Convolution library used in the upcoming Zones Convolution plugin. Implements a non-uniform partitioned convolution (NUPC) scheme with modified Garcia optimal partitioning and time distributed transforms. This is able to run on a single thread without large spikes in load for variable block sizes without additional latency.

Latte is a convolutional neural network (CNN) inference engine written in C++ and uses AVX to vectorize operations. The engine runs on Windows 10, Linux and macOS Sierra.

A C++ library that produces images based on different luminosity profiles

2D and 3D Matrix Convolution and Matrix Multiplication with CUDA

GUI Image-Editor written in C++ and Qt. Provides functionalities as Editing and Kernel convolution.

fast 2D convolution implementation benchmark

Test framework and reference implementation of our algorithms relating to the real-time simulation of human vision.