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

Student version of Assignment 1 for Stanford CS336 - Language Modeling From Scratch

Synthetic data curation for post-training and structured data extraction

GraphGen: Enhancing Supervised Fine-Tuning for LLMs with Knowledge-Driven Synthetic Data Generation

Verify Precision of all Kimi K2 API Vendor

Intercept AI requests, track usage, inject MCP tools centrally

Generate a timeline of your day, automatically

claude code工具集合

[NeurIPS 2025 Spotlight] Reasoning Environments for Reinforcement Learning with Verifiable Rewards

一个用于在 macOS 上平滑你的鼠标滚动效果或单独设置滚动方向的小工具, 让你的滚轮爽如触控板 | A lightweight tool used to smooth scrolling and set scroll direction independently for your mouse on macOS

Training setup for Langchain's Open Deep Research

Environments for LLM Reinforcement Learning

YiZhao: A 2TB Open Financial Corpus. Data and tools for generating and inspecting YiZhao, a safe, high-quality, open-source bilingual financial corpus (Chinese and English).

An open-source, all-in-one AI API router

Weighs the soul of incoming HTTP requests to stop AI crawlers

1 min voice data can also be used to train a good TTS model! (few shot voice cloning)

🚀🚀 「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 docker registry proxy run on cloudflare worker.

A lightweight LMM-based Document Parsing Model

Route, manage, and analyze your LLM requests across multiple providers with a unified API interface.

Command-line tools for onedrive integration

AI模型聚合管理中转分发系统，支持将多种大模型转为统一格式调用，支持OpenAI、Claude、Gemini等格式，可供个人或者企业内部管理与分发渠道使用。🍥 The next-generation LLM gateway and AI asset management system supports multiple languages.

[CVPR 2025] A Comprehensive Benchmark for Document Parsing and Evaluation

Python logging made (stupidly) simple

Debug, evaluate, and monitor your LLM applications, RAG systems, and agentic workflows with comprehensive tracing, automated evaluations, and production-ready dashboards.

Model Context Protocol Servers

The SOTA Open-Source Browser Agent for autonomously performing complex tasks on the web