Disaggregated serving system for Large Language Models (LLMs).

A tiny yet powerful LLM inference system tailored for researching purpose. vLLM-equivalent performance with only 2k lines of code (2% of vLLM).

Kernel-Bypass LibOS Architecture

计算机自学指南

GPU-accelerated vector query processing system that supports large vector datasets beyond GPU memory.

Automatic resource configuration for serverless workflows.

SGLang is yet another fast serving framework for large language models and vision language models.

A global object store with S3 interface that optimize performance and cost

阅读3服务器版，桌面端，iOS可用。后端 Kotlin + Spring Boot + Vert.x + Coroutine ；前端 Vue.js + Element。麻烦点点star，关注一下公众号【假装大佬】❗️ Demo服务器由于未备案已被关停，建议自行搭建

Artifacts for our SIGCOMM'23 paper Ditto

Examples of matplotlib codes and plots

Inference code for Llama models

My homepage

Artifacts for our NSDI'23 paper TGS

北京大学学位论文Latex模板(for Mac)，基于pkuthss v1.9.1修改，主要进行Mac的适配修改、个人最佳实践说明、配置踩坑经验

Vector search with bounded performance.

Legado 3.0 Book Reader with powerful controls & full functions❤️阅读3.0, 阅读是一款可以自定义来源阅读网络内容的工具，为广大网络文学爱好者提供一种方便、快捷舒适的试读体验。

PKU LaTeX

NVIDIA Linux open GPU kernel module source

⏰ Collaboratively track deadlines of conferences recommended by CCF (Website, Python Cli, Wechat Applet) / If you find it useful, please star this project, thanks~

Wechat Chat History Exporter 微信聊天记录导出备份程序

A cloud-native vector database, storage for next generation AI applications

Open deep learning compiler stack for cpu, gpu and specialized accelerators

PipeSwitch: Fast Pipelined Context Switching for Deep Learning Applications