VISIONARY❌ILLUSTRIOUS→NEW⭕ERA
⟦ ✦ CЯΞΛŦЇVΞ CHΞCKPӨЇNŦ ✦ ⟧
≋ where ⌬ Anime ✧ 3D ✧ Art ⌬ collide ≋
⟪ ⚙ Steps: 28 ⚙ ⟫ ≫
⟪ ✦ CFG: 3.5 ✦ ⟫ ≫
⟪ ⌬ Sampler: Euler a ⌬ ⟫
VISIONARY✦ILLUSTRIOUS:
visions... the beginning
Creative checkpoint - Where Anime, 3D, and Art collide
Pain:
What happens when even the darkness turns its back on you?
With a slight touch of anime influence, carrying the unique charm of New Fantasy Core
Surreal Chaos:
Do you find your way in a broken world?
I have merged this checkpoint with zero matching checkpoints... can you master it?
FEAR:
Fear teaches... did you come to learn or to hide?
huge merge of 16 checkpoints
📢 Feedback & Support
💬 Share your results or ask questions
✨ Suggestions for improvements? Always welcome!
🧠 Workflow in images
👍 Like
🔹 Enjoy generating! 🔹
Description
FAQ
Comments (8)
very good
This is by far my favorite diffusion model. Thank you. I struggle with IP adapter and reactor on this model. Any help or a workflow to get me targeted characters?
Hey, while I really like this model is there any reliable way to make it not give characters extra limbs and stuff? Cause the majority of characters I generate tend to have a bunch of extra arms, legs, and even heads, even when I'm using negative prompts to tell it to stop. This shouldn't be an issue with my prompt as it only happens with this model, if I use any of the other models I have it works fine.
Which version are we talking about?
@VisionaryAI_Studio The Illustrious version.
@Persson civitai is doing crazy things in the comments right now. I hope that comes through.
this is actually the first time I've heard of this problem.
I've generated over 100,000 images with it myself, basically my starting point for every Lora, and I don't have these problems.
Are you using the right CFG and steps?
CFG 3.5, STEP:28
Sampler: Eula a (simple/normal/beta)
Unfortunately, I can't help you with that because I haven't heard of anyone else having this problem :(
@Persson You can, of course, use controllnet with open pose, which would solve the problem.
@VisionaryAI_Studio Yeah those are the CFG, Steps, and Sampler I use, so not sure what the issue is. As for using ControlNet, I know I can do that, I was just hoping for a solution that didn't require either finding pose references or generating my own.
Details
Available On (2 platforms)
Same model published on other platforms. May have additional downloads or version variants.