By downloading, you agree to the CreativeML Open RAIL-M
This is an all-around model for anime to semi-realistic.
Pros:
Great results with few words
Multiple subjects
Intense bright colors
Favors symmetry
Can use Danbooru tags, long sentences, or both
Works well with embeddings and hypernetworks
Cons:
No photorealism
Poses can be tricky to summon correctly
Long heavily emphasized prompts are penalized
Sexual acts almost always lead to a Cronenberg fever dream
Tips:
To get intense colors, use (monochrome) in the negativ with psychedelic in the positiv
Highly recommending playing around with clip skip
VAE used : AnythingV3
By downloading, you agree to any licenses listed below.
Models used:
Anything-v3.0
EimisAnimeDiffusion_1.0v
ACertainModel
F222_v1
Dreamlike-diffusion-1.0 (license)
Description
FAQ
Comments (4)
This looks very interesting! I love the choices on models for this mix. Will be trying it out soon.
Hi! This model looks beautiful! Except every image I generate has like a filter that dims the intense colors. Am I doing something wrong? I downloaded the safetensors model.
Are you using AnythingV3 VAE ?
I missed that, thanks for the advice! Pictures are gorgeous!
Details
Files
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.








