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

A framework for few-shot evaluation of language models.

s1: Simple test-time scaling

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

A suite of image and video neural tokenizers

This repo contains the code for 1D tokenizer and generator

SEED-Voken: A Series of Powerful Visual Tokenizers

Losslessly encode text natively with arithmetic coding and HuggingFace Transformers

Octo is a transformer-based robot policy trained on a diverse mix of 800k robot trajectories.

Reinforcement Learning Environments for Omniverse Isaac Gym

A collaboration friendly studio for NeRFs

Clean PyTorch implementations of imitation learning algorithms

Framework for Easily Invertible Architectures

PyTorch3D is FAIR's library of reusable components for deep learning with 3D data

An open source robotics benchmark for meta- and multi-task reinforcement learning

Pre-training Reusable Representations for Robotic Manipulation Using Diverse Human Video Data

A Python package for intrinsic dimension estimation

MuJoCo is a physics engine for detailed, efficient rigid body simulations with contacts. mujoco-py allows using MuJoCo from Python 3.

DrQ: Data regularized Q

A Simulation Environment to train Robots in Large Realistic Interactive Scenes

PyBullet Planning