Stars
[EMNLP2025] "LightRAG: Simple and Fast Retrieval-Augmented Generation"
结合LightRAG 知识库的知识图谱智能体平台。LangChain v1 + Vue + FastAPI。集成主流大模型、LightRAG、MinerU、PP-Structure、Neo4j 、联网检索、工具调用。
verl: Volcano Engine Reinforcement Learning for LLMs
拼好RAG：手搓并融合了GraphRAG、LightRAG、Neo4j-llm-graph-builder进行知识图谱构建以及搜索；整合DeepSearch技术实现私域RAG的推理；自制针对GraphRAG的评估框架| Integrate GraphRAG, LightRAG, and Neo4j-llm-graph-builder for knowledge graph construct…
MedicalGPT: Training Your Own Medical GPT Model with ChatGPT Training Pipeline. 训练医疗大模型，实现了包括增量预训练(PT)、有监督微调(SFT)、RLHF、DPO、ORPO、GRPO。
《Agentic Design Patterns》中文翻译版
An implementation of a multimodal RAG system provides support for images, tables, and formulas.
💫 Toolkit to help you get started with Spec-Driven Development
AgentScope: Agent-Oriented Programming for Building LLM Applications
微舆：人人可用的多Agent舆情分析助手，打破信息茧房，还原舆情原貌，预测未来走向，辅助决策！从0实现，不依赖任何框架。
🐙 Guides, papers, lessons, notebooks and resources for prompt engineering, context engineering, RAG, and AI Agents.
🚀 「大模型」1小时从0训练26M参数的视觉多模态VLM！🌏 Train a 26M-parameter VLM from scratch in just 1 hours!
A light llama-like llm inference framework based on the triton kernel.
Gensis is a lightweight deep learning framework written from scratch in Python, with Triton as its backend for high-performance computing.
📚 《从零开始构建智能体》——从零开始的智能体原理与实践教程
🚀🚀 「大模型」2小时完全从0训练26M的小参数GPT！🌏 Train a 26M-parameter GPT from scratch in just 2h!
This is the homepage of a new book entitled "Mathematical Foundations of Reinforcement Learning."
Clean, Robust, and Unified PyTorch implementation of popular Deep Reinforcement Learning (DRL) algorithms (Q-learning, Duel DDQN, PER, C51, Noisy DQN, PPO, DDPG, TD3, SAC, ASL)
📚A curated list of Awesome LLM/VLM Inference Papers with Codes: Flash-Attention, Paged-Attention, WINT8/4, Parallelism, etc.🎉
本项目旨在分享大模型相关技术原理以及实战经验（大模型工程化、大模型应用落地）