AnimateDiff Legacy Animation v5.0 [ComfyUI]

Jerry Davos AI
15 May 202406:00

TLDRIn this tutorial video titled 'AnimateDiff Legacy Animation v5.0 [ComfyUI]', the creator demonstrates how to animate using ComfyUI and Anime Diff workflows. The process involves setting up inputs, using a model like 'mune anime', and adding fire effects. The video guides through batch size selection, frame rendering, and upscaling with specific settings. It also covers the use of the IP adapter for face fixing in animations, emphasizing the importance of matching FPS for smooth video speed. The tutorial concludes with a showcase of the final animation and a note of gratitude to Patreon supporters for making free educational content possible.


  • πŸ˜€ The tutorial is about creating animations using ComfyUI and AnimeDiff workflows.
  • 🎨 The video begins by instructing to drag and drop the first workflow to start the animation process.
  • πŸ” It mentions different components like inputs, animate, props, control, and settings as part of the workflow.
  • πŸ”§ The tutorial covers how to set up the output folder path for rendering frames and choosing the dimension and batch size.
  • 🌟 It introduces the use of a specific anime model, 'Mune Anime' with a character named 'Concept Pyromancer Laura' for fire effects.
  • πŸ“ The script explains how to choose an anime diff model for prompts and control net settings.
  • πŸ‘€ It discusses the use of a directory for open pose reference images and how to unmute and enable control net and open pose.
  • πŸŽ₯ The tutorial includes instructions on setting the FPS for the exporting video and rendering the queue.
  • πŸ“ˆ The process of upscaling the video is explained, including setting the output path, model settings, and upscale value.
  • πŸ–ΌοΈ The script also covers the use of a video face fixer workflow to improve the details of the faces in the animation.
  • πŸ”— The importance of adjusting the FPS according to the video speed is highlighted for both the upscaling and face fixing workflows.
  • πŸŽ‰ The tutorial concludes by mentioning that more workflow tutorials and other content are available for free on Patreon, thanks to the support of patrons.

Q & A

  • What is the title of the tutorial video?

    -The title of the tutorial video is 'AnimateDiff Legacy Animation v5.0 [ComfyUI].'

  • What software or tools are mentioned in the video for creating animations?

    -The video mentions ComfyUI and AnimeD as the tools for creating animations.

  • What are the main components of the workflow described in the video?

    -The main components of the workflow are inputs, animateD, prps, control, net, case sampler, settings, and video export.

  • What is the purpose of the 'Directory Group' in the workflow?

    -The 'Directory Group' is used to specify the directory for open pose reference images, which can be extracted from old renders or using the CN passes extractor workflow.

  • What is the batch size used in the tutorial for rendering the output?

    -The batch size used in the tutorial for rendering the output is 72.

  • Which anime model is chosen in the tutorial for adding fire effects?

    -The tutorial uses the 'mune anime' model and chooses 'concept pyromancer, Laura' to add fire effects.

  • What is the FPS set for exporting the video in the tutorial?

    -The FPS (frames per second) set for exporting the video in the tutorial is 12.

  • What is the purpose of the 'upscaling workflow' mentioned in the video?

    -The 'upscaling workflow' is used to increase the resolution of the video, making it clearer and more detailed.

  • What is the target resolution set in the upscaling workflow?

    -The target resolution set in the upscaling workflow is 1200.

  • What is the final step in the video after upscaling the video?

    -The final step after upscaling the video is to use the 'video2video face fixer workflow' to enhance the details of the faces in the video.

  • How does the video tutorial help the viewers with their AI artworks?

    -The video tutorial helps viewers by teaching them how to use ComfyUI and AnimeD workflows to create animations, offering insights on how to add effects, control the rendering process, upscale videos, and fix faces for better detail.



🎨 'Animating with Comfy UI and Anime,D Workflow'

This paragraph outlines the process of creating an animation using Comfy UI and Anime,D workflows. The tutorial begins by guiding users to set up their first workflow with inputs, animation, properties, and controls. It then introduces a batch or single operation option, a case sampler, and video export settings. The user is instructed to copy and paste the output folder path, choose the output dimension, and set the batch size. The tutorial specifies using a 'mune anime' model with a 'concept pyromancer, Laura' to add fire effects, adjusting the weight to 0.5. It also covers selecting an anime diff model for prompts and turning off the control net by default. The user is shown how to use a directory for open pose reference images, which can be extracted using a CN passes extractor. The paragraph concludes with instructions on rendering the queue and upscaling the video, adjusting the FPS for the exporting video, and using a video2video face fixer workflow to enhance the animation.


πŸŽ₯ 'Post-Production and Support Acknowledgement'

The second paragraph discusses post-production techniques such as frame interpolation for smoothness using Flow Frames. It also mentions that the creator posts workflow tutorials and other content on their Patreon for free, allowing everyone to learn and improve their AI artworks. The paragraph acknowledges the support of Patreons, emphasizing that their contributions are highly valued and are what keep the creator motivated to continue providing free tutorials.




AnimateDiff is a software tool used for generating animations and visual effects. In the context of the video, AnimateDiff is utilized to create a specific animation workflow, which is the main focus of the tutorial. The script mentions 'AnimateDiff Legacy Animation v5.0' indicating the version and legacy status, suggesting it's a traditional method being taught.


ComfyUI seems to refer to a user interface design that is easy and comfortable to use. Although not explicitly defined in the script, it implies that the software or tools being used have a user-friendly interface, which is important for the tutorial's audience to follow along easily.


In the video script, workflows refer to a series of steps or processes involved in creating an animation. The term is used to describe the sequence of operations that the user must follow using AnimateDiff and other tools to achieve the desired animation effects.


Inputs in this context are the initial data or materials required to start the animation process. The script mentions 'drag and drop the first workflow' which suggests that the user needs to input the starting point for the animation sequence.


Animation is the process of creating the illusion of motion in a sequence of images. The video is a tutorial on how to make an animation using specific software and techniques. The term is central to the video's theme as it is the end goal the tutorial aims to achieve.

πŸ’‘control net

A control net in the script refers to a feature within the animation software that allows for the manipulation and control of various aspects of the animation. It is mentioned that 'control net would be turned off by default,' indicating it's an optional tool that can be enabled for more precise control over the animation.

πŸ’‘case sampler

The case sampler appears to be a component of the software that allows the user to sample or select different cases or scenarios within the animation. It is part of the settings mentioned in the script, suggesting it's used to customize the animation's appearance or behavior.

πŸ’‘video export

Video export is the process of rendering the final animation into a video file that can be shared or used elsewhere. The script mentions 'video export settings' which implies configuring the parameters for how the animation will be saved as a video.


Upscaling in the context of the video refers to the process of increasing the resolution of the animation to make it higher quality. The script describes an 'upscaling workflow' which is used to enhance the visual quality of the rendered animation.

πŸ’‘face fixer

Face fixer is a term used in the script to describe a feature or process that improves or corrects the facial features in the animation. It is part of the final steps in the workflow where the facial details are refined to make the animation more realistic or visually appealing.

πŸ’‘frame interpolation

Frame interpolation is a technique used to create smooth transitions between frames in an animation. The script mentions adding 'frame interpolation for smoothness,' indicating that this technique is used to enhance the fluidity of the animation and make it appear more natural.


Learn to make an animation using ComfyUI and AnimeD workflows.

Link to the tutorial provided in the description.

Drag and drop the first workflow to start the animation process.

Explanation of the workflow components: inputs, animateD, prps, control.

Introduction of the control net with batch or single op options.

Use of case sampler settings in the animation.

Copying and pasting the output folder path for rendering frames.

Choosing the dimension and batch size for the output.

Selection of the 'mune anime' model and customization of its weight.

Adding fire effects to the animation with a specific model.

Disabling the control net by default and enabling it for directory use.

Utilization of open pose reference images for the animation.

Explanation on how to extract open pose images using CN passes extractor.

Adjusting the FPS for the exporting video to control the speed.

Rendering the queue and waiting for the animation to process.

Moving on to the upscaling workflow for video enhancement.

Details on inputting video and settings for the upscaling workflow.

Selection of model settings and prompts for the upscale value.

Copying the video path and setting the load cap for rendering.

Setting the target resolution and adjusting the FPS for the video.

Initiating the render process and waiting for the output.

Using the video2video face fixer workflow for detailed facial features.

Instructions on setting up the video2video face fixer workflow.

Adding prompts for more detailed faces and upscaling for better quality.

Starting the render for the face fix and observing the outcome.

Final touches with frame interpolation for smoothness.

Acknowledgment of Patreon supporters and the importance of their support.

Invitation to learn more and improve AI artworks through Patreon tutorials.