赛博苏神：你的 research 好老师

Muon in Int8 Precision Made Possible

Compositional Muon release

DFlash: Block Diffusion for Flash Speculative Decoding

Build academic conference posters as a single HTML/CSS file, rendered to print-ready PDF via headless Chromium. A coding-agent skill.

Official repository for Parallax (Parameterized Local Linear Attention)

[MLSys 26] 🥇 Solution for Gated Delta Net Track of MLSys 26 Flash infer competition

PolyGLU: a drop-in replacement for SwiGLU in transformer FFN blocks. Inspired by neurotransmitter-receptor diversity, each neuron routes between 4 qualitatively distinct activation functions via hy…

The official implementation of the W&B Models and Weave MCP server.

Official PyTorch Implementation of Gated DeltaNet-2: Decoupling Erase and Write in Linear Attention

Efficient Infinite Context Transformers with Infini-attention Pytorch Implementation + QwenMoE Implementation + Training Script + 1M context keypass retrieval

Skills for writing tilelang and debugging with CUDA toolkits.

MNBVC(Massive Never-ending BT Vast Chinese corpus)超大规模中文语料集。对标chatGPT训练的40T数据。MNBVC数据集不但包括主流文化，也包括各个小众文化甚至火星文的数据。MNBVC数据集包括新闻、作文、小说、书籍、杂志、论文、台词、帖子、wiki、古诗、歌词、商品介绍、笑话、糗事、聊天记录等一切形式的纯文本中文数据。

high-performance linear attention kernel library built on TileLang

Compact and Agent-Native MoE Training System

Claude Code skill that removes signs of AI-generated writing from text

An agentic skills framework & software development methodology that works.

Mobile and Web client for Codex and Claude Code, with realtime voice, encryption and fully featured

提取微信聊天记录，将其导出成HTML、Word、CSV文档永久保存，对聊天记录进行分析生成年度聊天报告

Python tool for converting files and office documents to Markdown.

A unified library of SOTA model optimization techniques like quantization, distillation, pruning, neural architecture search, speculative decoding, etc. It compresses deep learning models for downs…

Understand and test language model architectures on synthetic tasks.

The Newton-Muon optimizer

FlashInfer: Kernel Library for LLM Serving