Using ComfyUI Cloud

EchoMimic_v2

EchoMimic_v2 can voice - drive digital humans, supporting gestures. It optimizes animations with Audio - Pose strategy, etc. Custom full - body animations are available: upload images, audio, and gesture videos to generate coordinated ones. Widely used for digital human live - streaming and virtual anchors, it boosts vividness and immersion.

2.2K

2024.11.21

MemoAvatar

MemoAvatar builds a voice - action mapping model with advanced algorithms. By learning from huge amounts of relevant data, it can analyze voice features and transform them into digital human's motions, expressions, and lip - sync, achieving full - scale voice - driven digital humans.

3.2K

2024-12-18

CogVideo

CogVideoX-I2V uses deep learning techniques like GAN and VAE to interpret input static images as feature representations with rich semantic information. After learning from lots of video data, it grasps the dynamic change rules of image elements in the time dimension and can generate coherent and logical video content from a single image.

1.2K

2024-12-24

RH_OminiControl

ComfyUI_RH_OminiControl utilizes the computational power of the RTX 4090 GPU by splitting the pipeline loading and uses the schnell model to generate spatial and fill effects to improve efficiency. It relies on commonly used libraries and supports custom model paths.

1.8K

2024-12-20

FLUX-PuLID

The basic version v0.9.1 of Flux Pulid has been updated, with a large number of optimizations made to the nodes. The image generation nodes have improved the image resolution, details and the effect of color transitions. The model invocation nodes have optimized the algorithm, sped up the loading speed and enhanced the computational efficiency. The data processing nodes have strengthened the compatibility with different data formats.

0.7K

2024-10-05

FLUX-Redux

FLUX-Redux is an adapter for image variation generation for FLUX.1 base models. Input an image, it can generate a slightly different copy to refine the original image, integrate into complex workflows for image restyling, and restyle based on text via API with an image and a language prompt.

2.5K

2024-11-13

FLUX-Fill

Flux Fill integrates advanced inpainting & outpainting models. It edits and expands real/AI - generated images per text descriptions and masks, simplifies modification, blends new content naturally, and enhances the creative experience.

4.4K

2024-12-10

SUPIR

ComfyUI-SUPIR is a SUPIR upscaling wrapper node for ComfyUI. After the update, the functions of the node have been optimized. It supports loading CLIP models from SDXL checkpoints, improves memory management, and offers more sampling options.

1.7K

2024-11-29