My learning notes for ML SYS.

A SOTA quantization algorithm for high-accuracy low-bit LLM inference, seamlessly optimized for CPU/XPU/CUDA, with multi-datatype support and full compatibility with vLLM, SGLang, and Transformers.

depyf is a tool to help you understand and adapt to PyTorch compiler torch.compile.

This is a Chinese translation of the CUDA programming guide

bev_lane_det with lower resolution

『ゼロから作る Deep Learning ❸』(O'Reilly Japan, 2020)

Sample codes for my CUDA programming book

Nano vLLM

Implement a ChatGPT-like LLM in PyTorch from scratch, step by step

Disaggregated serving system for Large Language Models (LLMs).

A quick command line tool for OS X to get your geographic coordinates using the CoreLocation framework.

Model Context Protocol Servers

ONNX deployment of darknet (YOLOv3/YOLOv4)

Extend OpenRLHF to support LMM RL training for reproduction of DeepSeek-R1 on multimodal tasks.

FlashMLA: Efficient Multi-head Latent Attention Kernels

HIPIFY: Convert CUDA to Portable C++ Code

LangGPT: Empowering everyone to become a prompt expert! 🚀 📌 结构化提示词（Structured Prompt）提出者 📌 元提示词（Meta-Prompt）发起者 📌 最流行的提示词落地范式 | Language of GPT The pioneering framework for structured & meta-prompt…

The official GitHub page for the survey paper "A Survey of Large Language Models".

本项目是一个面向小白开发者的大模型应用开发教程，在线阅读地址：https://datawhalechina.github.io/llm-universe/

stay tuned.

中文羊驼大模型三期项目 (Chinese Llama-3 LLMs) developed from Meta Llama 3

A lightweight multilingual LLM

《Template Metaprogramming with C++ 》的非专业个人翻译

《C++ Templates The Complete Guide - second edition》的非专业个人翻译

AISystem 主要是指AI系统，包括AI芯片、AI编译器、AI推理和训练框架等AI全栈底层技术