Fast Multi-dimensional Sparse Attention

A library for high performance deep learning inference on NVIDIA GPUs.

Fast implementation of BERT inference directly on NVIDIA (CUDA, CUBLAS) and Intel MKL

Perplexity GPU Kernels

Geometric Augmentation for Text Image

Running BERT without Padding

Efficient framework-agnostic data loading

Microsoft Collective Communication Library

Kolosal AI is an OpenSource and Lightweight alternative to LM Studio to run LLMs 100% offline on your device.

Emoji 😄 for c++ developers 👍

torchcomms: a modern PyTorch communications API

An acceleration library that supports arbitrary bit-width combinatorial quantization operations

terashuf shuffles multi-terabyte text files using limited memory

C++ model train&inference framework

Learning Compact Binary Descriptors with Unsupervised Deep Neural Networks (CVPR16)

Search Formula-1——A distributed high performance massive data engine for enterprise/vertical search

ncnn和pnnx格式编辑器

Fast compressed trie dictionary library

Now this project has been changed to https://github.com/topcpporg/rest_rpc

Command-line utility for multipattern search using liblightgrep

Simple area to test out concepts or try to reproduce isolated issues in Drake.

A control program for the robot TARS constructed for a bachelor thesis at Chalmers university of technology.

Computation using data flow graphs for scalable machine learning