[ICML 2025] Official PyTorch implementation of "FlatQuant: Flatness Matters for LLM Quantization"

Optimized primitives for collective multi-GPU communication

Meta Lingua: a lean, efficient, and easy-to-hack codebase to research LLMs.

Tile-Based Runtime for Ultra-Low-Latency LLM Inference

d3LLM: Ultra-Fast Diffusion LLM 🚀

A next.js web application that integrates AI capabilities with draw.io diagrams. This app allows you to create, modify, and enhance diagrams through natural language commands and AI-assisted visual…

Fast, memory-efficient attention column reduction (e.g., sum, mean)

Paper Debugger is the best overleaf companion

The official implementation of Mantis: A Versatile Vision-Language-Action Model with Disentangled Visual Foresight

ShinkaEvolve: Towards Open-Ended and Sample-Efficient Program Evolution

Dexbotic: Open-Source Vision-Language-Action Toolbox

Accelerating Large-Scale Reasoning Model Inference with Sparse Self-Speculative Decoding

Running VLA at 30Hz frame rate and 480Hz trajectory frequency

🤗 PEFT: State-of-the-art Parameter-Efficient Fine-Tuning.

🤗 LeRobot: Making AI for Robotics more accessible with end-to-end learning

Real-Time VLAs via Future-state-aware Asynchronous Inference.

Build, evaluate and train General Multi-Agent Assistance with ease

A framework for efficient model inference with omni-modality models

A NCCL extension library, designed to efficiently offload GPU memory allocated by the NCCL communication library.

Single-file implementation to advance vision-language-action (VLA) models with reinforcement learning.

Vortex: A Flexible and Efficient Sparse Attention Framework

⚙️ All-in-One menu bar app, hide 💻MacBook Pro's notch, dark mode, AirPods, Shortcuts

A set of vim, zsh, git, and tmux configuration files.

TypeScript AI AI Function Calling Framework enhanced by compiler skills.

[COLM 2025] Official repository for R2E-Gym: Procedural Environment Generation and Hybrid Verifiers for Scaling Open-Weights SWE Agents

DSPy: The framework for programming—not prompting—language models

High Performance KV Cache Store for LLM

[ASPLOS'26] Taming the Long-Tail: Efficient Reasoning RL Training with Adaptive Drafter

A curated list of Diffusion Model in RL resources (continually updated)