Stars
Implement a ChatGPT-like LLM in PyTorch from scratch, step by step
gpt-oss-120b and gpt-oss-20b are two open-weight language models by OpenAI
Intelligent automation and multi-agent orchestration for Claude Code
An AI agent development platform with all-in-one visual tools, simplifying agent creation, debugging, and deployment like never before. Coze your way to AI Agent creation.
A General-purpose Task-parallel Programming System in C++
一款简单易用和高性能的AI部署框架 | An Easy-to-Use and High-Performance AI Deployment Framework
🚀🚀 「大模型」2小时完全从0训练64M的小参数GPT！🌏 Train a 64M-parameter GPT from scratch in just 2h!
Lightweight coding agent that runs in your terminal
SkyReels-V2: Infinite-length Film Generative model
The AI Scientist-v2: Workshop-Level Automated Scientific Discovery via Agentic Tree Search
Qwen3-Coder is the code version of Qwen3, the large language model series developed by Qwen team.
[SIGGRAPH Asia 2025] DreamO: A Unified Framework for Image Customization
AI-powered reverse engineering assistant that bridges IDA Pro with language models through MCP.
Open Source DeepWiki: AI-Powered Wiki Generator for GitHub/Gitlab/Bitbucket Repositories. Join the discord: https://discord.gg/gMwThUMeme
Qwen3 is the large language model series developed by Qwen team, Alibaba Cloud.
Technical report of Kimina-Prover Preview.
[CVPR 2025 Best Paper Nomination] FoundationStereo: Zero-Shot Stereo Matching
Implementing DeepSeek R1's GRPO algorithm from scratch
《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
Simple operating system in C++, written from scratch
[ICLR 2025] OpenVid-1M: A Large-Scale High-Quality Dataset for Text-to-video Generation
Lets make video diffusion practical!
PyTorch native quantization and sparsity for training and inference