Cosmos-Predict1 is a collection of general-purpose world foundation models for Physical AI that can be fine-tuned into customized world models for downstream applications.

High-speed Large Language Model Serving for Local Deployment

[CVPR 2026] Official repo for "EVATok: Adaptive Length Video Tokenization for Efficient Visual Autoregressive Generation"

[ICLR 2026] LongLive: Real-time Interactive Long Video Generation

Official repository for LTX-Video

[CVPR2024 Highlight] VBench - We Evaluate Video Generation

HunyuanVideo: A Systematic Framework For Large Video Generation Model

Wan: Open and Advanced Large-Scale Video Generative Models

An agentic skills framework & software development methodology that works.

An all-in-one enhancement suite for Google Gemini & AI Studio - timeline navigation, folder management, prompt library, and chat export in one powerful extension. / Google Gemini & AI Studio 全能增强插件…

An ns-3 library to simulate the real-time communication (RTC) application.

Retrieve author and publication information from Google Scholar in a friendly, Pythonic way without having to worry about CAPTCHAs!

不知道哪些基质该留哪些该扔？逝逝这个终末地基质妙妙小工具吧！

A simple demo repo for tuning sqlite with MLOS.

Codebase for Autothrottle (NSDI 2024)

Implementation, benchmarking, and proofs for the FRCC congestion control algorithm (NSDI26)

Here are my personal paper reading notes (including machine learning systems, AI infrastructure, and other interesting stuffs).

cluster data collected from production clusters in Alibaba for cluster management research

反重力Agent代理一键脚本，支持WSL、SSH远程

The web front end for Overleaf, a web-based collaborative LaTeX editor. Regular pull updates from upstream. DO NOT Fork, just cherry-pick the first commit.

Tools and scripts for reproducing results for FMCAD 2024 paper on Neural Lyapunov Barrier Certificates

Open-source benchmark suite for cloud microservices

A fast and user-transparent parallel simulator implementation for ns-3

ITU-T Rec. P.1203 Implementation

Source code for the paper ' Achieving Fairness Generalizability for Learning-based Congestion Control with Jury'

SelfTune is an RL framework that enables systems and service developers to automatically tune various configuration parameters and other heuristics in their codebase, rather than manually-tweaking,…