Stars
DeepInfer: High-performance, large-scale inference framework for DeepSeek models
Minimalistic large language model 3D-parallelism training
A practical guide to building enterprise AI agents across their full lifecycle.
learn LLM inference system on Apple Silicon for systems engineers: build a tiny vLLM + Qwen
learn database internals by building a storage engine in Rust
Build LLM agents and multi-agent systems from scratch, with MCP, Skills, and A2A
Lightweight coding agent that runs in your terminal
🚀 An open-source, hands-on curriculum bridging the gap from basic RL concepts to LLM alignment, RLVR, and advanced Agentic systems.
Mini Hermes: From Zero to Self-Improving AI Agent in Python
《开源大模型食用指南》针对中国宝宝量身打造的基于Linux环境快速微调（全参数/Lora）、部署国内外开源大模型（LLM）/多模态大模型（MLLM）教程
本地 LLM 微调实战课（SFT / LoRA / PEFT / DPO / QLoRA）+ 浏览器训练工作室可视化 loss/tensor/adapter | Hands-on local fine-tuning course with browser studio
Official repository of my book "A Hands-On Guide to Fine-Tuning LLMs with PyTorch and Hugging Face"
Hermes Agent 深度研究：从机制本质到代码实现，从 OpenClaw 渊源到 EvoMap 争议
Implement a reasoning LLM in PyTorch from scratch, step by step
The agent that grows with you
[Pytorch] Generative retrieval model using semantic IDs from "Recommender Systems with Generative Retrieval"
Use PEFT or Full-parameter to CPT/SFT/DPO/GRPO 600+ LLMs (Qwen3.8, DeepSeek-V4, GLM-5.1, InternLM3, Llama4, ...) and 300+ MLLMs (Qwen3-VL, Qwen3-Omni, InternVL3.5, Ovis2.5, GLM5.3, Gemma4, Llava, P…
Materials for the Hugging Face Diffusion Models Course
Demystify AI agents by building them yourself. Local LLMs, no black boxes, real understanding of function calling, memory, and ReAct patterns.
Build AI agents from first principles using a local LLM - no frameworks, no cloud APIs, no hidden reasoning.
The open-source visual AI programming environment and TypeScript library
[ECCV2024] IDM-VTON : Improving Diffusion Models for Authentic Virtual Try-on in the Wild
Open-source framework for the research and development of foundation models.
《大模型白盒子构建指南》：一个全手搓的Tiny-Universe