Visual Studio Code

A tutorial on RDMA based programming using code examples

Efficient and easy multi-instance LLM serving

Working Set Size tools

A small library to modify all page-table levels of all processes from user space for x86_64 and ARMv8.

Linux-based partitioning hypervisor

The easiest way to serve AI apps and models - Build Model Inference APIs, Job queues, LLM apps, Multi-model pipelines, and more!

Self-host LLMs with vLLM and BentoML

NVIDIA Linux open GPU kernel module source

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

Tensors and Dynamic neural networks in Python with strong GPU acceleration

Fast OS-level support for GPU checkpoint and restore

High-speed Large Language Model Serving for Local Deployment

Serverless LLM Serving for Everyone.

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

OpenZFS on Linux and FreeBSD

An open and reliable container runtime

Filesystem overlay for transparent, distributed migration of active data across separate storage systems.

SpotServe: Serving Generative Large Language Models on Preemptible Instances

NVIDIA device plugin for Kubernetes

This is a typora theme inspired by Vue document style. 一个类似于 Vue 文档风格的 Typora Markdown 编辑器主题。

CUDA checkpoint and restore utility

Checkpoint/Restore tool

cricket is a virtualization solution for GPUs

Solutions to all questions of the book Introduction to the Theory of Computation, 3rd edition by Michael Sipser

XRP: In-Kernel Storage Functions with eBPF