A minimal GPU design in Verilog to learn how GPUs work from the ground up

A fast, clean, responsive Hugo theme.

A simple light hugo theme with TailwindCSS that supports full responsive.

Render ASCII art as SVG diagrams

Wan: Open and Advanced Large-Scale Video Generative Models

verl: Volcano Engine Reinforcement Learning for LLMs

A neurosymbolic perspective on LLMs

A course on aligning smol models.

Dion optimizer algorithm

Shoegaze rock-inspired Turing-complete fractal music language. 1st place @ MIT Media Lab hackathon.

GRPO training code which scales to 32xH100s for long horizon terminal/coding tasks. Base agent is now the top Qwen3 agent on Stanford's TerminalBench leaderboard.

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

Hierarchical Reasoning Model Official Release

Simple and efficient DeepSeek V3 SFT using pipeline parallel and expert parallel, with both FP8 and BF16 trainings

Democratizing Reinforcement Learning for LLMs

Text-audio foundation model from Boson AI

Improve your resumes with Resume Matcher. Get insights, keyword suggestions and tune your resumes to job descriptions.

Quickly get started with EVI and Next.js

Open-source framework for the research and development of foundation models.

AI wearables. Put it on, speak, transcribe, automatically

AI Crash Course to help busy builders catch up to the public frontier of AI research in 2 weeks

Helpful tools and examples for working with flex-attention

Multi-LoRA inference server that scales to 1000s of fine-tuned LLMs

MLX: An array framework for Apple silicon

Free, local, open-source AI app builder ✨ v0 / lovable / Bolt alternative 🌟 Star if you like it!

Development repository for the Triton language and compiler

Ray is an AI compute engine. Ray consists of a core distributed runtime and a set of AI Libraries for accelerating ML workloads.

Distributed training framework for TensorFlow, Keras, PyTorch, and Apache MXNet.

Train transformer language models with reinforcement learning.