Starred repositories
REDMI K90 Pro Max / myron 国行的设备树
数字生命卡兹克开源的 AI Skills 合集 | Agent Skills: leader（帮你定义目标）, neat-freak 洁癖, hv-analysis, khazix-writer & more — Claude Code, Codex & 40+ agents
中国专利.skill：从项目文档到交底书编写（挖点·查新·脱敏成文）+ 专利通俗解读（叙事·图谱·Obsidian 私库）。
🧠「大模型」2小时完全从0训练64M的小参数LLM！Train a 64M-parameter LLM from scratch in just 2h!
CLI/GUI for managing the battery charging status for Apple silicon (M1, M2, M3) Macs
FineTune, a macOS menu bar app for per-app volume control, multi-device output, audio routing, and 10-band EQ. Free and open-source alternative to SoundSource.
OpenClaw-RL: Train any agent simply by talking
Proxifier Alternative to redirect any Windows/MacOS/Linux TCP and UDP traffic to HTTP/Socks5 proxy
Monitor your local neighbourhood's bluetooth activity
一键将 Markdown 和网页 AI 对话（ChatGPT/DeepSeek等）完美粘贴到 Word、WPS 和 Excel 的效率工具 | One-click paste Markdown and AI responses (ChatGPT/DeepSeek) into Word, WPS, and Excel perfectly.
Machine Learning Engineering Open Book
Autonomous GPU Kernel Generation & Optimization via Deep Agents
Scan the Hallucination Citation of Academic papers. Convert second-hand citation to official version
A high-performance RL training-inference weight synchronization framework, designed to enable second-level parameter updates from training to inference in RL workflows
RLinf: Reinforcement Learning Infrastructure for Embodied and Agentic AI
My learning notes for ML SYS.
注释的nano_vllm仓库，并且完成了MiniCPM4的适配以及注册新模型的功能
verl/HybridFlow: A Flexible and Efficient RL Post-Training Framework
Fast OS-level support for GPU checkpoint and restore
A library for accelerating Transformer models on NVIDIA GPUs, including using 8-bit and 4-bit floating point (FP8 and FP4) precision on Hopper, Ada and Blackwell GPUs, to provide better performance…
Pipeline Parallelism Emulation and Visualization
国家中小学智慧教育平台 电子课本下载工具，帮助您从智慧教育平台中获取电子课本的 PDF 文件网址并进行下载，让您更方便地获取课本内容。