ComfyUI-AniSora: Generate Diverse Anime Style Videos with One Click

ComfyUI-AniSora is an open-source anime video generation tool integrated with the ComfyUI platform. With the Index-AniSora model as its core, it is capable of generating video content in a variety of anime styles with a single click. Users can create videos in a variety of styles, including anime episodes, domestic original animation, manga adaptations, VTuber videos, anime promos, and funny parody videos, through simple operations. The tool is suitable for anime fans, content creators and developers. Its open source nature allows users to access the code for free and customize it according to their needs. It is easy to install and runs with the ComfyUI environment and AniSoraV2 model weights. The tool supports rapid generation of high-quality videos, is intuitive to operate, and is suitable for a variety of creative scenarios.

Function List

• Generate multiple anime style videos with one click, including anime episodes, domestic original animation, manga adaptations and more.
• Supports VTuber video and anime promo (PV) production to meet virtual anchor and promotion needs.
• Provide funny parody video (ghost animation) generation, suitable for entertainment content creation.
• Generate high quality animation videos based on Index-AniSora model.
• Open source code, users are free to modify and extend the function.
• Integrated ComfyUI platform with intuitive interface for beginners and professional users.

Using Help

Installation process

To use ComfyUI-AniSora, users need to prepare the ComfyUI environment and install relevant dependencies first. Below are the detailed installation steps:

1. Install ComfyUI
   Make sure you have ComfyUI installed on your computer. If not, you can clone the code from the official ComfyUI GitHub repository (https://github.com/comfyanonymous/ComfyUI) and follow the instructions to complete the installation. ComfyUI is a Python-based workflow tool. ComfyUI is a Python-based workflow tool, Python 3.8 or above is recommended.
2. Clone ComfyUI-AniSora Warehouse
   Open a terminal and go to ComfyUI's custom_nodes directory and then clone the ComfyUI-AniSora repository:

   cd ComfyUI/custom_nodes
   git clone https://github.com/Yuan-ManX/ComfyUI-AniSora.git
   

3. Installation of dependencies
   Go to the cloned ComfyUI-AniSora directory and install the required dependencies:

   cd ComfyUI-AniSora
   pip install -r req-fastvideo.txt
   pip install -r requirements.txt
   pip install -e .
   

   These commands install the Python libraries related to video generation and ensure that the tool works properly.

4. Download AniSoraV2 Model Weights
   ComfyUI-AniSora requires AniSoraV2 model weights to generate videos. Users need to visit the Hugging Face platform (the link is in the GitHub repository description) to download the model weights. Once downloaded, place the weights file in ComfyUI's model directory (usually the ComfyUI/models). Ensure that the filename and path match the configuration requirements of ComfyUI-AniSora.
5. Start ComfyUI
   Once the installation is complete, launch ComfyUI. in the ComfyUI interface, you will see that the nodes for ComfyUI-AniSora are integrated into the workflow.

Usage

ComfyUI-AniSora With ComfyUI's node-based workflow operation, users can drag and drop nodes to build a video generation process. Below are the detailed operation steps of the main functions:

1. Create anime style videos
   • Open the ComfyUI interface and find the ComfyUI-AniSora related node (usually starts with "AniSora") in the node library.
   • Add the AniSoraV2 model node and load the downloaded model weights.
   • Configure input parameters such as video style (anime episode, manga adaptation, etc.), resolution, frame rate, etc. The user can use text prompts to describe the desired image, for example, "a young girl in a kimono under a cherry blossom tree".
   • Connect the generation node and click "Run" to generate the video. The generation time depends on hardware performance and video length.
2. Generate VTuber Video
   • Choose a VTuber-style preset template (if available) or manually enter a prompt that describes the virtual anchor's image, such as "Virtual anchor with short blue hair talking on air".
   • Can be combined with audio input nodes to add voiceover or background music to enhance the video effect.
   • Preview the generated result, adjust the parameters and re-generate until you are satisfied.
3. Creating funny parody videos (ghost animation)
   • In the prompt, enter a funny element or a description of the object to be imitated, e.g. "cartoon character repeats dance moves".
   • Use loop nodes to control the number of times a video clip is repeated, generating a funny effect of looping playback.
   • After exporting the video, it can be further edited with other tools.
4. Customized Functions
   • Users can modify ComfyUI-AniSora's code to add new features, such as supporting more styles or optimizing generation speed.
   • In the GitHub repository's issues maybe pull requests page, where users can view updates to community contributions or suggest improvements.

caveat

• hardware requirement: Higher-performance GPUs are required to generate high-quality video, and NVIDIA graphics cards are recommended to accelerate processing.
• model weight: Ensure that the version of AniSoraV2 weights downloaded from Hugging Face is compatible with ComfyUI-AniSora.
• adjust components during testing: If the generation fails, check the log file to confirm that the dependencies are installed correctly or that the model weights are loaded correctly.
• Community Support: Problems can be found in the GitHub repository's issues page to ask questions, the community is more active and usually gets a quick response.

With these steps, users can easily get started with ComfyUI-AniSora and quickly generate high-quality anime videos. The node-based operation of the tool allows even beginners to get started quickly, while the open source feature provides more possibilities for developers.

application scenario

1. Anime and Manga Content Creation
   Anime fans can utilize ComfyUI-AniSora to create original animated shorts, such as making a fantasy-style adventure story video. Users just need to enter a description of the scene, and the tool generates a video clip that fits the anime style and is suitable for sharing on social media platforms.
2. VTuber Video Production
   Virtual anchors can use the tool to quickly generate live background or dynamic image videos. For example, generate an opening animation of a virtual anchor in a sci-fi city background to enhance the professional feel of the live content.
3. Manga Adaptation Video
   Comic book authors can convert static comics into dynamic videos. For example, key scenes in the comic are inputted into cues to generate dynamic battle scenes to enhance the presentation of the work.
4. Funny Video Creation
   Content creators can produce funny parody videos (ghost animation), for example, combining popular terriers with anime style to generate looping funny clips to attract viewers' attention.

QA

1. Is ComfyUI-AniSora free?
   Yes, ComfyUI-AniSora is an open source tool and users can download the code and instructions for use from GitHub for free. However, you need to prepare the ComfyUI environment and AniSoraV2 model weights by yourself.
2. What hardware is required to run it?
   A computer with an NVIDIA GPU with at least 8GB of video memory is recommended to ensure efficient video generation.CPU operation may be slow and ineffective.
3. How do I fix a model weight loading failure?
   Verify that the path to the weights file is correct and check that you downloaded the correct version of AniSoraV2. If this still fails, you can find the correct version of AniSoraV2 in the GitHub issues Page to search for a solution or ask a question.
4. Is it possible to generate non-anime style videos?
   ComfyUI-AniSora is mainly for anime style videos. If you need other styles, you need to modify the code or use other ComfyUI plugins.