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

A SCIP indexer for Ruby, powered by Sorbet

Fast, Flexible and Portable Structured Generation

The Modular Platform (includes MAX & Mojo)

A generative world for general-purpose robotics & embodied AI learning.

Competitive GPU kernel optimization platform.

MLX: An array framework for Apple silicon

Verified Rust for low-level systems code

Unified framework for robot learning built on NVIDIA Isaac Sim

CUDA Python: Performance meets Productivity

Diagram as Code Tool Written in Rust with Draggable Editing

Lightweight static analysis for many languages. Find bug variants with patterns that look like source code.

SCIP indexer for the C# programming language. Built with the Roslyn .NET compiler

AcadHomepage: A Modern and Responsive Academic Personal Homepage

Buttercup finds and patches software vulnerabilities

Fast and memory-efficient exact attention

Time zone database and code

Run your own AI cluster at home with everyday devices 📱💻 🖥️⌚

How to ensure correctness and ship LLM generated kernels in PyTorch

Training PyTorch models with differential privacy

[ICLR 2025 Spotlight] MetaUrban: An Embodied AI Simulation Platform for Urban Micromobility

Official Website for ACM ICPC at UCLA

1st place. Team Crane's solution the 9/6 Jane Street x GPU_MODE hackaton! @aswinkumar1999 @rathull @kyolebu

Mamba SSM architecture

A tiny vLLM-inspired optimized inference server.

Examples in the MLX framework

JAX-based neural network library

Code for the 9/6 Hackathon

Fine-Tuning Vision-Language-Action Models: Optimizing Speed and Success