A high-throughput and memory-efficient inference and serving engine for LLMs

Lightweight Linux namespace sandbox with persistent shell and instant reset. Designed for RL training

A scrollable-tiling Wayland compositor.

Open-source, secure environment with real-world tools for enterprise-grade agents.

slime is an LLM post-training framework for RL Scaling.

Zhihu on Obsidian | 知乎 Obsidian 插件

EditorJumper is a JetBrains IDE plugin that allows you to seamlessly jump between JetBrains IDE and other popular code editors (such as VS Code, Cursor, Trae, and Windsurf). It maintains your curso…

🚀 Efficient implementations for emerging model architectures

Programmer's guide about how to cook at home.

Awesome-Long2short-on-LRMs is a collection of state-of-the-art, novel, exciting long2short methods on large reasoning models. It contains papers, codes, datasets, evaluations, and analyses.

FlashMLA: Efficient Multi-head Latent Attention Kernels

Code for "Critique Fine-Tuning: Learning to Critique is More Effective than Learning to Imitate" [COLM 2025]

A toolbox for benchmarking trustworthiness of multimodal large language models (MultiTrust, NeurIPS 2024 Track Datasets and Benchmarks)

tiktoken is a fast BPE tokeniser for use with OpenAI's models.

小红书笔记 | 评论爬虫、抖音视频 | 评论爬虫、快手视频 | 评论爬虫、B 站视频 ｜ 评论爬虫、微博帖子 ｜ 评论爬虫、百度贴吧帖子 ｜ 百度贴吧评论回复爬虫 | 知乎问答文章｜评论爬虫

ChatGPT - Review & Rebuttal: A browser extension for generating reviews and rebuttals, powered by ChatGPT. 利用 ChatGPT 生成审稿意见和回复的浏览器插件

some code of numerical optimization method

Creating Pretty HTML From R Markdown

人工智能结课作业（A星八数码/广度优先/深度优先/粒子群寻优算法/遗传算法/蚁群算法/BP神经网络/卷积神经网络）

Implement Machine learning algorithm by myself using Python 3.6

编程之路