[CVPR 2025 Oral]Infinity ∞ : Scaling Bitwise AutoRegressive Modeling for High-Resolution Image Synthesis

"E-RayZer: Self-supervised 3D Reconstruction as Spatial Visual Pre-training" official implementation.

Curate, Annotate, and Manage Your Data in LightlyStudio.

PyTorch media decoding and encoding

Matplotlib styles for scientific plotting

The most powerful and modular diffusion model GUI, api and backend with a graph/nodes interface.

code based for rectified flow

A python library for self-supervised learning on images.

Official repository for "VideoPrism: A Foundational Visual Encoder for Video Understanding" (ICML 2024)

Interactive Post-Training for Vision-Language-Action Models

official training and inference code of bitwise tokenizer

Normalized Transformer (nGPT)

Benchmarking Knowledge Transfer in Lifelong Robot Learning

Code for ICCV'2025 (Best student paper honorable mention) "RayZer: A Self-supervised Large View Synthesis Model"

MAGI-1: Autoregressive Video Generation at Scale

State-of-the-art Image & Video CLIP, Multimodal Large Language Models, and More!

Official PyTorch implementation of One-Minute Video Generation with Test-Time Training

Mastering Diverse Domains through World Models

NVIDIA Isaac GR00T N1.6 - A Foundation Model for Generalist Robots.

This package contains the original 2012 AlexNet code.

(CVPR 2025) From Slow Bidirectional to Fast Autoregressive Video Diffusion Models

JPEG XL image format reference implementation

FlashMLA: Efficient Multi-head Latent Attention Kernels

[arXiv: 2502.05178] QLIP: Text-Aligned Visual Tokenization Unifies Auto-Regressive Multimodal Understanding and Generation

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

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

This repo contains the code for 1D tokenizer and generator