Programmable datacenter-scale infrastructure for Agents.

GPU Functional Descriptor for memory access

TPU inference for vLLM, with unified JAX and PyTorch support.

一个基于 Andrej Karpathy 的 LLM Wiki 模式 实现的 Obsidian 知识库，利用 LLM 维护可复利的个人知识层。

The agent that grows with you

Implementation inspired by Andrej Karpathy's LLM Knowledge Base pattern.

构建受监督的、自我进化的 Agent 组织的基础设施 | Infrastructure for supervised, self-improving agent organization. 飞书/Telegram 手机端运行 Claude Code 或 Kimi Code（双引擎，两家原生订阅直接用），共享记忆、Agent 工厂、定时任务、通信总线。

An AI Agent for Automated JAX Pallas Kernel Optimization

Discord/Feishu AI Bot powered by Claude Code

Official implementation of S3Diff

Oh my tmux! My self-contained, pretty & versatile tmux configuration made with 💛🩷💙🖤❤️🤍

A curated list of awesome iOS ecosystem, including Objective-C and Swift Projects

Google 💚 MCP

The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.

Tokamax: A GPU and TPU kernel library.

A powerful set of tools for ComfyUI

TurboDiffusion: 100–200× Acceleration for Video Diffusion Models

A simple, performant, and scalable Jax LLM!

Home for "How To Scale Your Model", a short blog-style textbook about scaling LLMs on TPUs

🤗 Diffusers: State-of-the-art diffusion models for image, video, and audio generation in PyTorch and FLAX.

TPU inference for vLLM, with unified JAX and PyTorch support.