[SiggraphAsia25] OmnimatteZero: Fast Training-free Omnimatte with Pre-trained Video Diffusion Models

Qwen3-TTS is an open-source series of TTS models developed by the Qwen team at Alibaba Cloud, supporting stable, expressive, and streaming speech generation, free-form voice design, and vivid voice…

Official implementation of "VideoMaMa: Mask-Guided Video Matting via Generative Prior", CVPR 2026

High quality training free inpaint for every stable diffusion model. Supports ComfyUI

A ComfyUI node pack that implements FreeLong (NeurIPS 2024) spectral blending for Wan 2.2 video generation

[CVPR 2026] PersonaLive! : Expressive Portrait Image Animation for Live Streaming

Native and Compact Structured Latents for 3D Generation

A lightweight yet powerful audio-to-MIDI converter with pitch bend detection

Online MIDI Editor: signal

PainterFLF2V boosts first-last frame motion with inverse structural repulsion. PainterFLF2V: Dynamically enhances the original first-last frame node, allowing you to customize the dynamic enhanceme…

Real-Time Diffusion-Based Streaming Video Super-Resolution / 基于Diffusion架构的实时视频流超分模型

Multilingual Document Layout Parsing in a Single Vision-Language Model

ToonOut, a fork of BiRefNet focused on background removal for anime images. We open-source our dataset & our weights. See our paper at: https://arxiv.org/abs/2509.06839

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

Collections of web event spines made by HoYo

[CVPR'26] ObjectClear: Precise Object and Effect Removal with Adaptive Target-Aware Attention

A KCP packet sniffer + visualizer in one, backend rewritten in Go. (For SR)

OmniGen2: Exploration to Advanced Multimodal Generation. https://arxiv.org/abs/2506.18871

集中整理了TCAX论坛的287个py脚本，方便大家下载并用TCAX制作特效字幕。

Asset extraction tool for unity games, supports GI 6.0+, HSR 3.6+, ZZZ 2.2+ and many more (*ﾟ∀ﾟ*)

AssetStudioMod - modified version of Perfare's AssetStudio, mainly focused on UI optimization and some functionality enhancements.

MoviiGen 1.1: Towards Cinematic-Quality Video Generative Models

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

Ultralytics YOLO with Additional Knowledge Distillation Capability

[NeurIPS 2025] Image editing is worth a single LoRA! 0.1% training data for fantastic image editing! Surpasses GPT-4o in ID persistence~ MoE ckpt released! Only 4GB VRAM is enough to run!

3D x AI hybrid editor, built with three.js

UniAnimate-DiT: Human Image Animation with Large-Scale Video Diffusion Transformer