Enhance-A-Video: Better Generated Video for Free

Official repository of "SAMURAI: Adapting Segment Anything Model for Zero-Shot Visual Tracking with Motion-Aware Memory"

A suite of image and video neural tokenizers

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

Efficient vision foundation models for high-resolution generation and perception.

Official implementation of "Realistic Morphology-preserving Generative Modelling of the Brain"

This repo has all the basic things you'll need in-order to understand complete vision transformer architecture and its various implementations.

Train VAE like a boss

[CSUR] A Survey on Video Diffusion Models

A general fine-tuning kit geared toward diffusion models.

xDiT: A Scalable Inference Engine for Diffusion Transformers (DiTs) with Massive Parallelism

Official inference repo for FLUX.1 models

Efficient Triton Kernels for LLM Training

Course to get into Large Language Models (LLMs) with roadmaps and Colab notebooks.

Implementation of rectified flow and some of its followup research / improvements in Pytorch

code for "Diffusion Forcing: Next-token Prediction Meets Full-Sequence Diffusion"

Code for Adam-mini: Use Fewer Learning Rates To Gain More https://arxiv.org/abs/2406.16793

Official implementation for "Identifying and Solving Conditional Image Leakage in Image-to-Video Diffusion Model" (NeurIPS 2024)

UNet diffusion model in pure CUDA

A one-stop library to standardize the inference and evaluation of all the conditional video generation models.

[ICML 2024] EvTexture: Event-driven Texture Enhancement for Video Super-Resolution

LLM101n: Let's build a Storyteller

An open-source toolbox for fast sampling of diffusion models. Official implementations of our works published in ICML, NeurIPS, CVPR.

A comprehensive collection of KAN(Kolmogorov-Arnold Network)-related resources, including libraries, projects, tutorials, papers, and more, for researchers and developers in the Kolmogorov-Arnold N…

[NeurIPS 2024] CV-VAE: A Compatible Video VAE for Latent Generative Video Models

Train and fine-tune diffusion models. Perform image-to-image class transfer experiments.