A. Izzuddin Al Faruq

ComfyUI Unnofficial Implementation of Spectral Progressive Diffusion for Efficient Image and Video Generation for Anima

ComfyUI-Spectrum-SDXL is a custom node for ComfyUI that implements the Spectrum diffusion acceleration technique

parser to convert html to jsx

ComfyUI custom node Implements "Foveated Diffusion: Prioritized Distortion for Efficiency" to achieve 2–4× faster sampling via mixed-resolution tokenization and patched spatial attention.

Frequency-aware RoPE reference-attention control for Flux.2 in ComfyUI: tune reference influence from style transfer to stronger structural/content copying.

PyTorch implementation of JiT https://arxiv.org/abs/2511.13720

An ERP system for plant management, developed as a practical project inspired by the 'Sistem Enterprise' course.

A Chrome/Edge browser extension for uploading images directly to Szurubooru with right-click context menu.

A Distributed Attention Towards Linear Scalability for Ultra-Long Context, Heterogeneous Data Training

fork for implement twinflow

No description.

No description.

Expert skill for AI agents to work on ComfyUI's backend architecture

No description.

No description.

Repository to host plugin related issues, and requests for LNReader

No description.

AI skills for generating ITS-format academic reports, built for agentic workflows.

No description.

headless renderer and web based editor for alight motion project

Autoresearch for GPU kernels. Give it any PyTorch model, go to sleep, wake up to optimized Triton kernels.

Final Project DLH

Training-free style transfer for DiT models.

Vibecoding practice

URL Shortener app built while exploring OpenCode

Accelerate inference in Flux and Sana for ComfyUI.

Tugas ingest-store matkul DLH brought by claude via antigravity

No description.

No description.

5026241093 Abdurrahman Izzuddin Al Faruq

Nodes for image juxtaposition for Flux in ComfyUI

No description.

X-Scrape: Autonomous Artist Account Detection using VLM

DFlash: Block Diffusion for Flash Speculative Decoding

Official implementation of Tuna-2: Pixel Embeddings Beat Vision Encoders for Unified Understanding and Generation

Game Engine for Idle Heroes like game

ComfyUI DyPE, enabling artifact-free 4K+ image generation for Qwen, Flux + Nunchacku

No description.

An experimental implementation of a node that implements Pixelrush (Anima) technique in Comfyui

[ICCV 2025] FreeFlux: Understanding and Exploiting Layer-Specific Roles in RoPE-Based MMDiT for Versatile Image Editing

Official JAX implementation of Riemannian Mean Flow for flow map learning on manifolds.

5026241093 Abdurrahman Izzuddin Al Faruq

Effort to port SeaCache to ComfyUI

Cache-DiT Node for Comfyui

No description.

[CVPR2026] Official repository for “DeCo: Frequency-Decoupled Pixel Diffusion for End-to-End Image Generation”

dLLM: Simple Diffusion Language Modeling

Flow matching for accelerated simulation of atomic transport

No description.

No description.

Train generative models with pytorch lightning

Official repository for “PixelGen: Pixel Diffusion Beats Latent Diffusion with Perceptual Loss”

🎨 A Style is Worth One Code: Unlocking Code-to-Style Image Generation with Discrete Style Space

An end‑to‑end web application optimized for serving the SDXL model efficiently. Designed with performance and scalability in mind, it provides a streamlined interface for deploying, managing, and delivering high‑quality outputs from SDXL with minimal overhead.

The simplest, fastest repository for training/finetuning medium-sized GPTs.

Taming large-scale few-step training with self-adversarial flows! 👏🏻

[NeurIPS 2025] U-REPA: Aligning Diffusion U-Nets to ViTs

No description.

No description.

No description.

No description.

(NeurIPS 2024 Oral 🔥) Improved Distribution Matching Distillation for Fast Image Synthesis

No description.

[AAAI 2026] VisionReward: Fine-Grained Multi-Dimensional Human Preference Learning for Image and Video Generation

No description.

No description.

No description.

No description.

Pytorch implementation of MeanFlow on ImageNet and CIFAR10

No description.

No description.

Transition Models: Rethinking the Generative Learning Objective

Official PyTorch Implementation of "Latent Diffusion Model Without Variational Autoencoder".

[ICLR'25 Oral] Representation Alignment for Generation: Training Diffusion Transformers Is Easier Than You Think

An official implementation of "CapRL: Stimulating Dense Image Caption Capabilities via Reinforcement Learning"

Official implementation of Decoupled MeanFlow

[NeurIPS'25 Spotlight] Boosting Generative Image Modeling via Joint Image-Feature Synthesis

No description.

No description.

No description.

Official PyTorch Implementation of "Diffusion Transformers with Representation Autoencoders"

Official PyTorch Implementation of "Scalable Diffusion Models with Transformers"

No description.

LLM based autonomous agent that conducts deep local and web research on any topic and generates a long report with citations.

No description.

Home Made Diffusion Models

Tongyi Deep Research, the Leading Open-source Deep Research Agent

Calligrapher: Freestyle Text Image Customization

Official code for "RB-Modulation: Training-Free Personalization of Diffusion Models using Stochastic Optimal Control"

No description.

Implementation of the Mesh-VQVAE of "VQ-HPS: Human Pose and Shape Estimation in a Vector-Quantized Latent Space" - ECCV 2024

Qwen-Image is a powerful image generation foundation model capable of complex text rendering and precise image editing.

[RSS 2023] Diffusion Policy Visuomotor Policy Learning via Action Diffusion

(CVPR 2025) Switti: Designing Scale-Wise Transformers for Text-to-Image Synthesis

Exploring Diffusion Transformer Designs via Grafting

PyTorch implementation for "Parallel Sampling of Diffusion Models", NeurIPS 2023 Spotlight

Accepted as [NeurIPS 2024] Spotlight Presentation Paper

Convolutional Neural Networks to predict the aesthetic and technical quality of images.

[NeurIPS 2023] Scaling Riemannian Diffusion Models

This is the official repository for the paper: "Energy Matching: Unifying Flow Matching and Energy-Based Models for Generative Modeling".

Contrastive unpaired image-to-image translation, faster and lighter training than cyclegan (ECCV 2020, in PyTorch)

fastdup is a powerful, free tool designed to rapidly generate valuable insights from image and video datasets. It helps enhance the quality of both images and labels, while significantly reducing data operation costs, all with unmatched scalability.

Code for Energy-Based Transformers paper

Image board engine, Danbooru-style. Tailored for text to image dataset

No description.

PDE-Transformer is a neural network architecture designed to efficiently process and predict the evolution of physical systems described by partial differential equations.

JoyCaption is an image captioning Visual Language Model (VLM) being built from the ground up as a free, open, and uncensored model for the community to use in training Diffusion models.

Official implementation of "XVerse: Consistent Multi-Subject Control of Identity and Semantic Attributes via DiT Modulation".

[CVPR 2025] MASt3R-SLAM: Real-Time Dense SLAM with 3D Reconstruction Priors

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

Multimodal Representation Alignment for Image Generation: Text-Image Interleaved Control Is Easier Than You Think!

A PyTorch library for implementing flow matching algorithms, featuring continuous and discrete flow matching implementations. It includes practical examples for both text and image modalities.

Official implementation of the paper: RICO: Improving Accuracy and Completeness in Image Recaptioning via Visual Reconstruction

Creates an index of images, queries a local LLM and adds tags to the image metadata

PyTorch Implementation of Diffusion Schrodinger Bridge Matching

No description.

OmniGen2: Unified Image Understanding and Generation.

No description.

A summary of related works about flow matching, stochastic interpolants

Easily turn large sets of image urls to an image dataset. Can download, resize and package 100M urls in 20h on one machine.

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

[Preprint] Efficient Generative Model Training via Embedded Representation Warmup

[CVPR 2025 Best Paper Award] VGGT: Visual Geometry Grounded Transformer

Open-Sora: Democratizing Efficient Video Production for All

[ICML 2025] The Diffusion Duality

A curated list of awesome papers on dataset distillation and related applications.

Elucidating the Design Space of Diffusion-Based Generative Models (EDM)

EDM2 and Autoguidance -- Official PyTorch implementation

[Preprint] GMem: A Modular Approach for Ultra-Efficient Generative Models

[CVPR2025 Highlight] Video Generation Foundation Models: https://saiyan-world.github.io/goku/

No description.

Native-resolution diffusion Transformer

maximal update parametrization (µP)

Consistency Distilled Diff VAE

FreeU: Free Lunch in Diffusion U-Net (CVPR2024 Oral)

Tiny AutoEncoder for Stable Diffusion

[ECCV 2024] HiDiffusion: Increases the resolution and speed of your diffusion model by only adding a single line of code!

PyTorch implementation of DiffMoE, TC-DiT, EC-DiT and Dense DiT

[CVPR 2025 Oral & Best Paper Award Candidate] Reconstruction vs. Generation: Taming Optimization Dilemma in Latent Diffusion Models

DDT: Decoupled Diffusion Transformer

Repository for Meta Chameleon, a mixed-modal early-fusion foundation model from FAIR.

No description.

AI based multi-label girl image classification system, implemented by using TensorFlow.

galeri gambar berbasis URL untuk pengumpulan dataset menggunakan PHP

[ICML'25] EQ-VAE: Equivariance Regularized Latent Space for Improved Generative Image Modeling.

Quantized Attention achieves speedup of 2-5x and 3-11x compared to FlashAttention and xformers, without lossing end-to-end metrics across language, image, and video models.

No description.

Official code for "Style Aligned Image Generation via Shared Attention"

Official implementation for "EGSDE: Unpaired Image-to-Image Translation via Energy-Guided Stochastic Differential Equations" (NIPS 2022)

[CVPR 2025] DiG: Scalable and Efficient Diffusion Models with Gated Linear Attention

No description.

MMaDA - Open-Sourced Multimodal Large Diffusion Language Models

Official implementations for paper: VACE: All-in-One Video Creation and Editing

No description.

🚀 Efficient implementations of state-of-the-art linear attention models in Torch and Triton

Official repo for: SuperEdit - Rectifying and Facilitating Supervision for Instruction-Based Image Editing

Official implementation of the paper: REPA-E: Unlocking VAE for End-to-End Tuning of Latent Diffusion Transformers

No description.

DreamO: A Unified Framework for Image Customization

This is an official implementation for "Swin Transformer: Hierarchical Vision Transformer using Shifted Windows".

[Preprint] UCGM: Unified Continuous Generative Models

Masked Diffusion Transformer is the SOTA for image synthesis. (ICCV 2023)

Scaling Diffusion Transformers with Mixture of Experts

No description.

利用AI大模型，一键生成高清短视频 Generate short videos with one click using AI LLM.

An official implementation of Flow-GRPO: Training Flow Matching Models via Online RL

Official repository for LTX-Video

Ming - facilitating advanced multimodal understanding and generation capabilities built upon the Ling LLM.

VARGPT-v1.1: Improve Visual Autoregressive Large Unified Model via Iterative Instruction Tuning and Reinforcement Learning

Agent framework and applications built upon Qwen>=3.0, featuring Function Calling, MCP, Code Interpreter, RAG, Chrome extension, etc.

Official inference repo for FLUX.1 models

No description.

Official inference framework for 1-bit LLMs

[ECCV2024] This is an official inference code of the paper "Glyph-ByT5: A Customized Text Encoder for Accurate Visual Text Rendering" and "Glyph-ByT5-v2: A Strong Aesthetic Baseline for Accurate Multilingual Visual Text Rendering""

[NeurIPS 2024] Official implementation of "Faster Diffusion: Rethinking the Role of UNet Encoder in Diffusion Models"

No description.

Implementation of Imagen, Google's Text-to-Image Neural Network, in Pytorch

Pixel-Space Generative Models

No description.

[ICLR 2025] Autoregressive Video Generation without Vector Quantization

Official implementation of I2I-Mamba, an image-to-image translation model based on selective state spaces

Wan: Open and Advanced Large-Scale Video Generative Models

[MLSys 2024 Best Paper Award] AWQ: Activation-aware Weight Quantization for LLM Compression and Acceleration

Let us control diffusion models!

Structured state space sequence models

Mamba SSM architecture

Qwen2.5-VL is the multimodal large language model series developed by Qwen team, Alibaba Cloud.

Lora beYond Conventional methods, Other Rank adaptation Implementations for Stable diffusion.

Model code for inferencing T5

Implementation of aspect ratio bucketing for training generative image models as described in: https://blog.novelai.net/novelai-improvements-on-stable-diffusion-e10d38db82ac

Qwen2.5 is the large language model series developed by Qwen team, Alibaba Cloud.

No description.

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

This repository includes the official implementation of our paper "Beyond Next-Token: Next-X Prediction for Autoregressive Visual Generation"

A TTS model capable of generating ultra-realistic dialogue in one pass.

Generative Models by Stability AI

[NeurIPS 2024 Best Paper][GPT beats diffusion🔥] [scaling laws in visual generation📈] Official impl. of "Visual Autoregressive Modeling: Scalable Image Generation via Next-Scale Prediction". An *ultra-simple, user-friendly yet state-of-the-art* codebase for autoregressive image generation!

Official PyTorch Implementation of "SiT: Exploring Flow and Diffusion-based Generative Models with Scalable Interpolant Transformers"

OmniMamba: Efficient and Unified Multimodal Understanding and Generation via State Space Models

Next-Token Prediction is All You Need