If you want to use more my checkpoint online generation, please visit here.
https://tensor.art/u/762555264535746522
V6
I adjusted the overall weight and reduced the color contrast.
Recommended Setting
Steps: 30
CFG scale: 7
Sampler: Euler a
ADetailer face_yolov8n/s.pt use can fix eyes
Positive Prompt
score_9, score_8_up, score_7_up, source_anime,
Negative Prompt
score_6,score_5,score_4,, worst quality, low quality, bad anatomy, bad hands, missing fingers, fewer digits, source_furry, source_pony, source_cartoon,3d, blurry,
v5
More complex compositions with richer colors.
v4
I have modified the coloring style, added some effects, optimized the overall composition, and corrected some body errors.
Recommended Setting no change
v3
Adjust the overall weights and add some materials.
The same settings as v2.
V2
Recommended Setting
Steps: 30
CFG scale: 7
Sampler: DPM++ 2M Karras OR Euler a
ADetailer face_yolov8n/s.pt use can fix eyes
Positive Prompt
score_9, score_8_up, score_7_up,rating_explicit,Negative Prompt
score_6,score_5,score_4,, worst quality, low quality, bad anatomy, bad hands, missing fingers, fewer digits, source_furry, source_pony, source_cartoon,3d, blurry, __________________________________________________________________________________________________
v1
Recommended Setting
Steps: 30
CFG scale: 7
Sampler: Euler a
ADetailer face_yolov8n/s use
Positive Prompt
score_9, score_8_up, score_7_up,Negative Prompt
score_6, score_5, score_4, source_cartoon,
3d, (censor),monochrome,blurry, lowres,watermark, Description
FAQ
Comments (5)
The only problem I have that the model likes to create multiple views of people, even when you type "multiple views" in the negatives. Is there a fix for that?
Can you provide a picture to post for me to see. But I'm not sure where the problem lies with the textual explanation. I didn't encounter this situation when I tested it myself.
@WAI0731 Its sometimes like a comic, where the same person is multiple times in one picture from different angles. Maybe it has something to do with the Loras I use.
This model is really incredible, I'm a fan of it, but it doesn't work well with the other Lora models. The images generated are more inspired by the Lora than the Checkpoint
You might be weighting the loras or the lora trigger words too heavily
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.