Visual node-based model merging tool for Stable Diffusion XL

Speak Friend and Enter

Repository for paper: Egalitarian Gradient Descent: A Simple Approach to Accelerated Grokking

Custom node to load models in INT8 for 1.5~2X Speed gains on 30 series cards.

PolarGrad: A Class of Matrix-Gradient Optimizers from a Unifying Preconditioning Perspective

Fast kernel library for Diffusion inference with multiple compute backends.

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

Fork of the official DF11 ComfyUI custom node that aims to support other model architectures, and support loading existing DF11 models that were originally only compatible with Diffusers. Also, a n…

DFloat11 [NeurIPS '25]: Lossless Compression of LLMs and DiTs for Efficient GPU Inference

A stochastic approximation method with block-coordinate optimal step-sizes (BCOS).

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

a vulkan post processing layer

A Vulkan Post Processing Layer For Linux. Next Gen vkBasalt. (CONTRIBUTORS NEEDED)

Fastest, Most Efficient, Minimal, Chunked & Target Quality Video Encoding

RAM is all you need

ComfyUI node implementing caching optimizations for Chroma

A no-nonsense SVT-AV1 encoder fork with sensible defaults and Quality of Life improvements.

Portable file server with accelerated resumable uploads, dedup, WebDAV, SFTP, FTP, TFTP, zeroconf, media indexer, thumbnails++ all in one file

A collection of random, experimental (and most likely "dum") samplers for ComfyUI

a vulkan post processing layer for linux

Fast, Numerically Stable, and Auto-Differentiable Spectral Clipping via Newton-Schulz Iteration

Lossless Scaling Frame Generation on Linux

ComfyUI implemtation for NAG

A collection of niche / personally useful PyTorch optimizers with modified code.

Official implementation of "Normalized Attention Guidance"

The official code that integrates MagCache (Fast Video Generation with Magnitude-Aware Cache) with ComfyUI.

A Ray Tracing-Inspired Approach to Neural Network Optimization

A single-line modification to any (dualizer-based) optimizer that allows the optimizer to adapt to the scale of the gradients as they change during training