As a long-time user of WebUI Forge and Reforge, it was pretty hard for me to switch to Anima, which only worked properly through ComfyUI. The workflows I used were functional but unfamiliar and scattered.
So I spent a lot of time building a unified workflow that mirrors the Forge pipeline I'm used to, integrating TXT2IMG, IMG2IMG and Inpaint so model settings, Loras and prompts are preserved.
So — I present my own workflow for Anima. I hope it helps those who are used to working in Automatic1111 and Forge. It may look complex, but I left notes with explanations. All links to the required models are included. Of course you’ll need to install a bunch of custom nodes, but what can you do — it’s Comfy 🤷♂️
Features:
🔹 Text to Image (+Face Detailer)
🔹 Image to Image
🔹 Inpaint (Whole picture & Only masked mode)
🔹 Upscale (+SeedVR +FaceDetailer)
Leave your questions, requests, and feedback. More features will be added in future versions. I hope this will at least slightly make life easier for Anima users in Comfy 🙂
Description
Have all basic features in one workflow. Face Detailer for one main face.
FAQ
Comments (8)
got excited to try this but it seems that the node "comfyui-prompt-reader-node" or SD Prompt Reader doesnt seem to be installing correctly no matter what i do the folder "stable_diffusion_prompt_reader" is empty no matter how i try to install it, tryed comfy, git, cmd, and zip.
im using comfyui desktop
update, i installed from https://github.com/receyuki/stable-diffusion-prompt-reader.git and wile now it does have that folders contents, its still not working for me.
for now i'll give the rest of the setup a go. thanks for posting it!
@bioguts01 Sorry that you have trouble installing this one. Maybe you need to update Comfy? Try to remove it from custom_nodes folder and reinstall with Manager. Anyway, you can bypass it or remove completely, its not that necessary. Thank you for your feedback!
У меня тоже не работала "comfyui-prompt-reader-node". Решается просто, вставьте в powershell:
$node = '\ComfyUI\custom_nodes\comfyui-prompt-reader-node'
$submodule = "$node\stable_diffusion_prompt_reader"
git clone https://github.com/receyuki/stable-diffusion-prompt-reader.git $submodule
git -C $submodule checkout 1a499becb0a88fd28ac3e4e09bd8917ce95c9629
@K0hler Thank you! that worked for me
Forge Neo does actually support Anima. Its updated basically every 2 days.
Still, from your description this sounds really good! I'll be sure to bookmark this for when i finally work up the courage to try out Comfy, because it is quite intimidating after a few years off A1111/Forge xD
Okay update after getting though the hurdles of comfyUI itself, the workflow is really nice and is making me like comfy more than A1111, the Lora loader you have in it is great. not too familiar with the different nodes and how to build a workflow myself but ended up switching the model loader with "Civitai Visual Browser Nodes by LX" might be a better one but it works for the visual aspect.
side note if you plan to update this in the future the anzhcfaceseg model for the face detailer seems to be taken down from multiple sites
Thanks for making and posting this workflow
Thank you very much for your feedback! In future versions I plan to add nsfw detailers and speed up generation. I'll take note on face detailer model