TurboDiffusion: 100–200× Acceleration for Video Diffusion Models

Information hub for our project training the largest possible historical LLMs.

Open-source platform to build and deploy AI agent workflows.

Domain-specific language designed to streamline the development of high-performance GPU/CPU/Accelerators kernels

📚A curated list of Awesome LLM/VLM Inference Papers with Codes: Flash-Attention, Paged-Attention, WINT8/4, Parallelism, etc.🎉

Vibe Workflow Platform for Non-technical Creators.

Tile-Based Runtime for Ultra-Low-Latency LLM Inference

Materials for learning SGLang

SGLang is a fast serving framework for large language models and vision language models.

DeepSpeed is a deep learning optimization library that makes distributed training and inference easy, efficient, and effective.

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。

How I Scaled from Zero to a Million Store on Dukaan, Without a CS Degree. .. A System Design Handbook by Subhash Choudhary

A lightweight sandboxing tool for enforcing filesystem and network restrictions on arbitrary processes at the OS level, without requiring a container.

All-in-One Sandbox for AI Agents that combines Browser, Shell, File, MCP and VSCode Server in a single Docker container.

The best ChatGPT that $100 can buy.

MCP for xiaohongshu.com

A Python library for extracting structured information from unstructured text using LLMs with precise source grounding and interactive visualization.

Supercharge Your LLM with the Fastest KV Cache Layer

Reproducing R1 for Code with Reliable Rewards

A Flexible Framework for Experiencing Heterogeneous LLM Inference/Fine-tune Optimizations

本仓库包含对 Claude Code v1.0.33 进行逆向工程的完整研究和分析资料。包括对混淆源代码的深度技术分析、系统架构文档，以及重构 Claude Code agent 系统的实现蓝图。主要发现包括实时 Steering 机制、多 Agent 架构、智能上下文管理和工具执行管道。该项目为理解现代 AI agent 系统设计和实现提供技术参考。

elephant.ai provides a Go backend and Next.js dashboard built around a shared Think → Act → Observe loop. The alex CLI/TUI, HTTP + SSE server, and web UI run the same runtime so operators and autom…

An open-source AI agent that brings the power of Gemini directly into your terminal.

🚀🚀 「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!

A collection of AI research papers that are well-organized and accessible for easy reference. It’s intended as a resource for anyone looking to explore key research in AI at their own pace.

🤗 PEFT: State-of-the-art Parameter-Efficient Fine-Tuning.

🌐 Make websites accessible for AI agents. Automate tasks online with ease.