10x faster matrix and vector operations

An efficient video loader for deep learning with smart shuffling that's super easy to digest

SeetaFace 2: open source, full stack face recognization toolkit.

This is the Official source code repository of Synfig Studio animation software

A high performance anime upscaler

Mirage Persistent Kernel: Compiling LLMs into a MegaKernel

Automatically Discovering Fast Parallelization Strategies for Distributed Deep Neural Network Training

SCUDA is a GPU over IP bridge allowing GPUs on remote machines to be attached to CPU-only machines.

Large-scale LLM inference engine

a fast and user-friendly runtime for transformer inference (Bert, Albert, GPT2, Decoders, etc) on CPU and GPU.

Maui Shell is a convergent shell for desktops, tablets, and phones.

FB (Facebook) + GEMM (General Matrix-Matrix Multiplication) - https://code.fb.com/ml-applications/fbgemm/

A primitive library for neural network

Trust & Safety tools for working together to fight digital harms.

Kids Is Data Stream

AutoPhrase: Automated Phrase Mining from Massive Text Corpora

A fast communication-overlapping library for tensor/expert parallelism on GPUs.

Benchmarking Deep Learning operations on different hardware

A flexible and efficient deep neural network (DNN) compiler that generates high-performance executable from a DNN model description.

Unsupervised text tokenizer focused on computational efficiency

A uniform interface to run deep learning models from multiple frameworks

PyTorch Extension Library of Optimized Graph Cluster Algorithms

An Efficient Lexical Analyzer for Chinese

Deploy your model with TensorRT quickly.

A performant and modular runtime for TensorFlow

The Munich Open-Source Large-Scale Multimedia Feature Extractor

Fast BPE

A repository for OCR, which inlcudes some classical OCR algorithms Pytorch implementation such as CTPN, EAST and CRNN.