A pipeline parallel training script for diffusion models.

A lightweight, powerful framework for multi-agent workflows

Fix AI pixel art and vector images right in your browser

Enjoy the magic of Diffusion models!

Official SeedVR2 Video Upscaler for ComfyUI

Hierarchical Reasoning Model Official Release

AI-generated pixel art often lacks true pixel precision, making it oversized or unsuitable for professional use. Pixel Perfect is designed to refine AI-generated pixel art, giving you greater contr…

Identity-Aware Tunneled Reverse Proxy Server with Dashboard UI

My custom (ethical) set of AI models to upscale textures and generate PBR maps

[CVPR 2024 Oral] Rethinking Inductive Biases for Surface Normal Estimation

Official repo of 𝗚𝗲𝗻𝗲𝗿𝗮𝘁𝗶𝘃𝗲 𝗠𝗼𝗱𝗲𝗹𝘀: 𝗪𝗵𝗮𝘁 𝗗𝗼 𝗧𝗵𝗲𝘆 𝗞𝗻𝗼𝘄? 𝗗𝗼 𝗧𝗵𝗲𝘆 𝗞𝗻𝗼𝘄 𝗧𝗵𝗶𝗻𝗴𝘀? 𝗟𝗲𝘁'𝘀 𝗙𝗶𝗻𝗱 𝗢𝘂𝘁! (Previously titled 𝘐𝘯𝘵𝘳𝘪𝘯𝘴𝘪𝘤 𝘓𝘰𝘙𝘈: 𝘈 𝘎𝘦𝘯𝘦𝘳𝘢𝘭𝘪𝘴𝘵 𝘈𝘱𝘱𝘳𝘰𝘢𝘤𝘩 𝘧𝘰𝘳 𝘋𝘪𝘴𝘤𝘰𝘷𝘦𝘳𝘪𝘯𝘨 𝘒𝘯𝘰𝘸𝘭𝘦𝘥𝘨𝘦 𝘪𝘯 𝘎𝘦𝘯𝘦𝘳𝘢𝘵𝘪𝘷𝘦 𝘔𝘰𝘥𝘦𝘭𝘴))

ComfyUI Plugin of Nunchaku

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

[NeurIPS 2025] OmniSVG is the first family of end-to-end multimodal SVG generators that leverage pre-trained Vision-Language Models (VLMs), capable of generating complex and detailed SVGs, from sim…

Lightweight coding agent that runs in your terminal

[ICLR2025 Spotlight] SVDQuant: Absorbing Outliers by Low-Rank Components for 4-Bit Diffusion Models

Simple go utility to download HuggingFace Models and Datasets

A simple, fast and user-friendly alternative to 'find'

One beautiful Ruby API for OpenAI, Anthropic, Gemini, Bedrock, OpenRouter, DeepSeek, Ollama, VertexAI, Perplexity, Mistral, GPUStack & OpenAI compatible APIs. Chat, Vision, Audio, PDF, Images, Embe…

The ultimate training toolkit for finetuning diffusion models

[SIGGRAPH Asia 2024, Journal Track] ToonCrafter: Generative Cartoon Interpolation

The fastest little YouTube web component on this side of the internet. The shadow dom web component version of Paul's lite-youtube-embed.

aider is AI pair programming in your terminal

Access large language models from the command-line

JS implementation of the xBR image scaler

modern color library for image processing and creative coding (oklab, xyz, p3, rec2020 + more)

isd (interactive systemd) – a better way to work with systemd units

Create GPU-rendered Svelte components

CVPR2025