A highly optimized LLM inference acceleration engine for Llama and its variants.

Official Implementation of EAGLE-1 and EAGLE-2

CUDA Templates for Linear Algebra Subroutines

TLLM_QMM strips the implementation of quantized kernels of Nvidia's TensorRT-LLM, removing NVInfer dependency and exposes ease of use Pytorch module. We modified the dequantation and weight preproc…

A high-throughput and memory-efficient inference and serving engine for LLMs

Efficient, Low-Resource, Distributed transformer implementation based on BMTrain

Fast and memory-efficient exact attention

Efficient Training (including pre-training and fine-tuning) for Big Models

Colossal-AI: A Unified Deep Learning System for Big Model Era

🤗 Transformers: State-of-the-art Machine Learning for Pytorch, TensorFlow, and JAX.

An Open Source Machine Learning Framework for Everyone

Bagua Speeds up PyTorch

A small TCP proxy written in Go

Third-person Survival Game for Unreal Engine 4 (Sample Project)

Using the famous cnn model in Pytorch, we run benchmarks on various gpu.

TensorFlow Tutorial and Examples for Beginners with Latest APIs

Prometheus Alertmanager

Embed the power of Lua into NGINX TCP/UDP servers

Python bindings for libgit2

Golang native MTR implementation

Vim-fork focused on extensibility and agility.

SendCloud mail API client in Go

The configuration framework for Zsh

Mars is a cross-platform network component developed by WeChat.

Network communication debugger iOS app Swift source code

职业信息文档

Transfer data using microphone/speaker on iOS devices

socket.io v0.7.2+ for iOS devices

Source code for the original TaskPaper for iOS.

Supervisor process control system for UNIX