⚡ Anima Workflow:
🛠️ Purpose & Design Philosophy
This workflow is designed for quality and autonomy, not speed. It follows an "all-in-one" philosophy: configure your settings, hit queue, and let the workflow handle everything from initial generation to high-res detailing in a single pass.
Not for Speed: If you want rapid-fire generations, this is not the tool for you. A solid, much faster alternative created by darksidewalker can be found here.
Personal Use: This was built for my personal production needs. It is not intended to be a "one-size-fits-all" solution, but I am sharing it for those who value the same high-fidelity results. Please adjust the settings to your preferences!
Heavy Duty: Due to the multi-stage processing, this workflow can be resource-intensive. In my experience, the detailers are not usually needed. YMMV.
On v1g and after: If the results are too blurry for you after USDU, you can try using the RTX nodes after it to help.
v2 is compatible with AIO versions of Anima models. Note: You might have to select a random model in the Checkpoint Loader node even if you are not using it. This is because it references your models/checkpoints folder. The opposite may also be the case if you are using the Checkpoint Loader node but not the Model Loader in the Diffusion Model Loader group which references the models/diffusion_models folder.
🚀 Key Features
Beyond standard generation and upscaling, this workflow integrates:
Power LoRA Loader: Efficiently manage multiple LoRAs without spaghetti wires.
Global Controls: Centralized Seed, Sampler, and Scheduler nodes for a unified experience.
Bypass Control to toggle features on/off.
Visual Validation: Integrated Image Comparer nodes to see exactly how your image evolves at every stage.
Upscaling: 2-stage upscaling using standard image upscaling and Ultimate SD Upscale (optional).
Triple Detailer Groups: 3-stage detailing using standard BBOX and SEGM detection models for faces, hands, and clothes.
CivitAI Ready: Images are saved with full metadata (Model, LoRAs, Prompts) for easy site parsing.
⚠️ Disclaimer & Compatibility
Install at Your Own Risk: Updating ComfyUI or adding custom nodes can break your environment. I am not responsible for any installation issues.
Portable Version: This was built and tested on the ComfyUI Portable version. Desktop app users may require additional troubleshooting.
"Your Mileage May Vary": Your environment is almost certainly different from mine. I do not guarantee 1:1 compatibility.
Nodes 2.0: I strongly recommend disabling Nodes 2.0. It causes unpredictable behavior; I will not provide support for any issues arising from its use.
🤝 Support & Boundaries
No DMs: Direct messages are disabled due to high volume. Please use the Discussions tab below. Check previous comments first (unless there aren't any yet), as most common questions may have already been answered.
Custom Requests: I do not take private requests for custom workflows. If you need a specific solution built, please post a Bounty on CivitAI. There are many talented creators here who will be happy to assist you for a fee.
Modifications: You are free to add or remove nodes as you see fit. However, if you change the internal logic, you are responsible for your own troubleshooting.
The only place I am actively maintaining this workflow is here on civitai. If my workflows are being posted and monetized elsewhere, whoever posted them is obligated to provide support.
Description
Not So Simple (or is it?) Anima Workflow - v1b Release Notes
This update focuses on streamlining the generation pipeline and optimizing for the Anima model by moving away from "flat" rescale tools and toward more dynamic detail enhancement.
Added: Anima Layer Replay Patcher (Anima Enhancer)
Purpose: Integrated the new Layer Replay Patcher to act as a dual-purpose detail enhancer and speed optimizer.
Implementation: Currently wired into the Initial Image Generation phase only.
Compatibility Note: Testing found that this node performed best at the base resolution; it has been intentionally omitted from the Ultimate SD Upscale (USDU) pass to maintain stability and prevent artifacts.
Documentation: I’ve added a Markdown note adjacent to the node within the workflow, summarizing the author’s recommended usage and settings.
Removed: Detail Daemon & RescaleCFG
Detail Daemon: Removed to prevent "node fighting." Even at low settings, Detail Daemon was found to conflict with the Layer Replay Patcher, leading to inconsistent results.
RescaleCFG: Discontinued to preserve textural depth. While RescaleCFG can help with color clamping, it was found to "flatten" complex textures like smoke and fine fabrics—which goes against the high-fidelity goals of this workflow.
Configuration & Defaults
Sample-Ready: Default values have been tuned to match the settings used in the latest showcase images (minus the specific prompt), providing a proven starting point for your own generations.
FAQ
Comments (17)
This is amazing G! Thank you o7
Another great workflow... i assume your images are a little blurry because of the image host compressing them.
They might be? Looks good to me locally at least lol.
open up one of them and just zoom in with the any browsers native +
Btw... 50 steps? How did you figured that so many steps will give a positive edge?
@Arararagi that's just me using the node from Anima Enhancer and based off the documentation from Anima suggesting up to 50 steps. I'm not some big brain behind settings.
If there is blur on the 50 step ones specifically it's because of using that node with Spectrum turned on. Plus, I don't care how something looks zoomed in. I just care about how it looks on my monitor. I'll take my results over hard edges and overly flat results any day.
Everyone has their own personal preferences and should use whatever they prefer. If I was doing this for selling to some target audience, then my approach would be different. However, as always, I am just doing this for myself and sharing my workflow since someone asked.
second detailer seems to break the chain and stop
hmm first I have heard of this, but will check now!
@sevora yep you're right. Looks like I was missing a connection on the "Use Loaded LORA Models?" switch. You can just add connect the checkpoint loader's model output to the false input on that toggle and it should work. I'll upload a fixed version of the workflow soon. Thanks for pointing it out.
Fixed v1b file uploaded. Was missing a model connection for the Detailer 2 group.
Thank you so much for fix
any chance to try img2img with anima model?
@sevora if you are asking if I will add IMG2IMG to this workflow:
I have no plans to add it since I don't use IMG2IMG. The same goes for ControlNet, regional prompting, and inpainting.
IMG2IMG is easy to setup, but not worth the hassle to explain to people who have never used it.
If you're seeing a bunch of these messages on the right side of your screen:
"Loading Error
A required resource failed to load. Please reload the page."
It looks like it's related to the Image Comparer nodes. This started happening to me a couple ComfyUI updates ago.
These can be safely ignored or if it triggers you that much, then you can just delete all the Image Comparer nodes from the workflow.
I might remove them in a future update, but as of this time, I plan on keeping them.
I'm not getting this error using the workflow on my end with the latest ComfyUI (nightly) and KJNodes 1.3.6 so the cause might be something else?
@mac2492 maybe, I even get similar errors on a blank workflow. Gonna try updating I guess. Hoping my Comfy install doesn't die in the process. (Edit: it died: time for a fresh install).
I did a bit of testing and ConditioningAverage doesn't seem to work with Anima? I tried Concat, Combine, Average, and string concat. All of them worked in different ways for Illustrious models, but for Anima only Combine and string concat actually worked as expected.
For your use case, Combine would probably be the closest alternative. String concat can't fail but wiring can get messy if anything adjusts conditioning (like an eventual IPAdapter equivalent). I've switched to string concat on my end since I use the USDU to make minor adjustments.
I'm actually going be dropping it from the workflow. It does work, but I don't use it enough to keep it.
Version 1c released.
Decided to work on this after breaking my comfy install (again).
I went the manual install route this time instead of the "portable" version.
10/10 don't recommend the 'manual' route if you like the ease of having pre-made .bat files for everything. When they say manual, they mean it!