gpt-oss-120b and gpt-oss-20b are two open-weight language models by OpenAI

Physics of Language Models: Part 4.2, Canon Layers at Scale where Synthetic Pretraining Resonates in Reality

Temporal Neural Networks

An AI agent system for solving International Mathematical Olympiad (IMO) problems using Google's Gemini, OpenAI, and XAI APIs.

Single-pass Adaptive Image Tokenization for Minimum Program Search | What's the Kolmogorov Complexity of an Image?

Muon is an optimizer for hidden layers in neural networks

Official Implementation of weights2weights

Scaling Properties of Diffusion Models For Perceptual Tasks (CVPR 2025)

Adaptive Length Image Tokenization via Recurrent Allocation | How many tokens is an image worth ?

The repository provides code for running inference with the Meta Segment Anything Model 2 (SAM 2), links for downloading the trained model checkpoints, and example notebooks that show how to use th…

Utilities for efficient fine-tuning, inference and evaluation of code generation models

Schedule-Free Optimization in PyTorch

Gaia Physics Engine

Synthesizing Moving People with 3D Control

Run PyTorch in JAX. 🤝

CUDA accelerated rasterization of gaussian splatting

Official codebase used to develop Vision Transformer, SigLIP, MLP-Mixer, LiT and more.

Official inference library for Mistral models

The official repo for [NeurIPS'22] "ViTPose: Simple Vision Transformer Baselines for Human Pose Estimation" and [TPAMI'23] "ViTPose++: Vision Transformer for Generic Body Pose Estimation"

Pytorch Implementation for Neural Point Characters (NPC)

Flax is a neural network library for JAX that is designed for flexibility.

DeepSpeed is a deep learning optimization library that makes distributed training and inference easy, efficient, and effective.

A minimal PyTorch Lightning OpenAI GPT w DeepSpeed Training!

Nerfbusters project

FAIR Sequence Modeling Toolkit 2

Generative Agents: Interactive Simulacra of Human Behavior

Inference code for Llama models

Code repository for the paper "On the Benefits of 3D Pose and Tracking for Human Action Recognition", (CVPR 2023)

Easily create large video dataset from video urls

An official code release of the paper RGB no more: Minimally Decoded JPEG Vision Transformers