Vidur: Microsoft's LLM Inference System Simulator
Vidur: Microsoft’s LLM Inference System Simulator Introduction Vidur is a high-fidelity LLM inference system simulator developed by Microsoft Research that enables engineers and researchers to study LLM deployment performance without requiring access to expensive GPU hardware. Published at MLSys’24 (paper...
Sniffnet: Open Source Network Traffic Monitor Built in Rust
Network traffic monitoring is essential for understanding what happens on your network – from identifying suspicious connections to tracking bandwidth usage across applications. Whether you are a system administrator watching for intrusions, a developer debugging API calls, or a privacy-conscious...
SGLang: The High-Performance LLM Serving Framework Powering 400K+ GPUs
SGLang is a fast and powerful LLM serving framework designed to deliver high-throughput, low-latency inference for large language models at production scale. Developed by the SGLang team at UC Berkeley and the open-source community, SGLang has rapidly become one of...
RAG-Anything: All-in-One RAG Framework for Multi-Modal Retrieval Augmented Ge...
Modern documents are no longer just walls of text. Research papers embed figures alongside equations, financial reports weave tables between narrative paragraphs, and technical documentation mixes diagrams with code snippets. Traditional retrieval augmented generation systems treat everything as plain text,...
Quarkdown: Markdown With Superpowers for Papers, Presentations, and More
Quarkdown: Markdown With Superpowers for Papers, Presentations, and More If you have ever wrestled with LaTeX to produce a research paper, or fought with slide decks that refuse to render correctly, Quarkdown is the project you have been waiting for....
Pixelle-Video: AI-Powered Fully Automated Short Video Engine
Pixelle-Video is an AI automated short video engine that transforms a simple text topic into a complete, polished video with zero manual editing. With over 7,200 stars on GitHub, this open-source project from AIDC-AI automates every step of short video...
Awesome Codex Skills: Curated Workflow Automation for OpenAI Codex CLI
OpenAI’s Codex CLI has rapidly become a go-to tool for developers who want AI-powered workflow automation directly from the terminal. The awesome-codex-skills repository by ComposioHQ delivers a curated collection of Codex skills that transform Codex from a code generator into...
ACE-Step UI: Open Source AI Music Generation Interface - The Suno Alternative
If you have been searching for an open source Suno alternative that delivers professional-grade AI music generation without subscription fees, ACE-Step UI is the answer. Built as a polished frontend for the ACE-Step 1.5 model, this project provides a Spotify-inspired...
InferSim: LLM Inference Simulation by Alibaba
Introduction InferSim is a lightweight, simulation framework developed by Alibaba’s Alimama AI Infra Team and Future Living Lab. It simulates the inference performance of Large Language Models (LLMs) on GPU hardware. In the context of simulation, InferSim bridges the gap...
wterm: A High-Performance Web Terminal Emulator Built with Zig and WASM
wterm: A High-Performance Web Terminal Emulator Built with Zig and WASM wterm (pronounced “dub-term”) is a terminal emulator for the web that renders directly to the DOM, giving you native text selection, copy/paste, browser find, and accessibility out of the...
Web Design Skill: AI Agent That Transforms Functional Web Pages Into Stunning...
Web Design Skill: AI Agent That Transforms Functional Web Pages Into Stunning Designs Modern LLMs can already produce functional web pages from simple prompts, but their output tends to converge on the same aesthetic: Inter font, blue primary buttons, purple-pink...
Video Use: Edit Videos With Coding Agents
Table of Contents Table of Contents What is Video Use? Why It Matters How It Works: Reading Video Through Text The Editing Pipeline Key Components transcribe.py / transcribe_batch.py pack_transcripts.py timeline_view.py render.py grade.py Hard Rules: Production Correctness Getting Started Prerequisites Installation...