Garrett Goon

Accelerating MoE with IO and Tile-aware Optimizations

No description provided.

🚀 Collection of components for development, training, tuning, and inference of foundation models leveraging PyTorch native components.

Notes on transformers models.

No description provided.

LLM Discovery via Co-Evolving Intrinsic World Model

AI-Driven Research Systems (ADRS)

SGLang is a high-performance serving framework for large language models and multimodal models.

An interface library for RL post training with environments.

An experimental implementation of compiler-driven automatic sharding of models across a given device mesh.

🤗 smolagents: a barebones library for agents that think in code.

A clean, modular SDK for building AI agents with OpenHands V1.

Democratizing Reinforcement Learning for LLMs

Claude Code is an agentic coding tool that lives in your terminal, understands your codebase, and helps you code faster by executing routine tasks, explaining complex code, and handling git workflows - all through natural language commands.

A version of verl to support diverse tool use

Composable transformations of Python+NumPy programs: differentiate, vectorize, JIT to GPU/TPU, and more

Our library for RL environments + evals

Async RL Training at Scale

verl: Volcano Engine Reinforcement Learning for LLMs

Ongoing research training transformer models at scale

SkyRL: A Modular Full-stack RL Library for LLMs

Typst package for creating scientific posters

An academic poster template for Typst

A place to store reusable transformer components of my own creation or found on the interwebs

Triton-based implementation of Sparse Mixture of Experts.

✨ AI Coding, Vim Style

No description provided.

Ship correct and fast LLM kernels to PyTorch

🚀 Efficient implementations of state-of-the-art linear attention models

PyTorch Single Controller