Sign In

QQQmix

Updated: Oct 5, 2024

animegirlsstyles

Download

1 variant available

fp16 SafeTensor

QQQmixV3.fp16.safetensors

Half precision, best balance (pruned) • 1.99 GB

Verified:

Type
Checkpoint Merge
Stats

525

Reviews
Published

Jul 12, 2023

Base Model

SD 1.5

Usage Tips
Clip Skip: 2
Hash
AutoV2
372EDE071C
default creator card background decoration
Followers - 161

161

Likes - 1355

1.4K

07305-1030767284-(highres, absurdres, madeon_1.1), (1girl_1.2), _poses_, _ftbUniforms_, BREAK _Tokyo, character focus, depth of field, intense an.png

NAI free......?

This model was only merged with the model considered not to contain NAI

If the model I used includes NAI, please let me know

Sample Image Settings

Dynamic thresholding

Mimic scale: 7 | Separate Feature Channels: True | Scaling Startpoint: MEAN | Variability Measure: AD | Interpolate Phi: 1 | Threshold percentile: 95 | Mimic mode: Half Cosine Up | Mimic scale minimum: 0 | CFG mode: Half Cosine Up | CFG scale minimum: 3.5 |

Used embeddings

EasyNegativeV2, negative_hand-neg

i2i Settings

Denoising strength:0.3

Tiled VAE(Mixture of Diffusers, 4x-AnimeSharp) + Tiled Diffusion + ControlNet(tile_colorfix+sharp)

CLIP

clip-vit-large-patch14-336

Source

v1.0

V2.0

I didn't like it, so I removed it.

V3.0

used LoRA trained on a model that is supposed to be free of NAI

V4.0

user : why not use SDXL?

my SD : "out of memory"