Huh. My guess would be that it has something to do with how image samples are loaded. Maybe certain colors appear differently when scaled down, but when shown in original resolution, it just appears normally or something.
Too much fun and wasted GPU resources for this thing. Decided to take on more challenging pose and concept. Same generation method as post #6321
masterpiece, best quality, (official art: 1.2), (mermaid with dark blue tail: 2), (lamia with light pink snake tail: 2), asymmetrical irises, 2girl, (loli: 2.3), (visible stomach: 1.2), looking at viewer, long hair, hair bow, tearing up, (wet clothes: 1.5), long tops, long sleeves, cheerleader skirt, (sportswear: 1.4), worries,🥹, blush, tired, (sitting on floor: 1.3), (locker room: 1.4), wooden floor, sun light, (sweating: 1.55), sparkle, on back, top down view, full body
Just noticed #mermaid tags only have 2 so far, so another one.
No idea how did AI came up with high contrast color, but I kinda liked it. Attempted to make it looks like it came from visual novel.
Same generation methods as post #6240 as Waifu for txt2img and NovelAI for img2img and inpainting.
masterpiece, best quality, (official art: 1.2), (mermaid: 2.2), (white fish tail: 1.4), (loli: 1.6), (visible stomach: 1.2), looking at viewer, long hair, hair bow, tearing up, (wet clothes: 1.5), long tops, long sleeves, beret hat, streetwear, wool jacket, visible stomach, 🥹, tired, (pouting: 1.5), (lying on table: 1.3), (coffee shop: 1.4), sun light, reflection, (sweating: 1.55), sparkle, on back, top down view, full body
Basically tons of inpainting and photoshop back and front (with multiple different prompt I forgot), with scaling then regenerated by set denoising strength to 0.3.
Used WaifuDiffusion (txt2img) for the base the NovelAI for majority of inpainting and img2img.
Attempted to make this looks like it came from a gacha game 😂
The main prompt is this (for both WaifuDiffusion and NovelAI):
masterpiece, best quality, (official art: 1.2), (mermaid: 1.9), (black fish tail: 1.4), (loli: 1.6), (visible stomach: 1.2), looking at viewer, magical, prism, (magical girl: 1.7), long hair, hair bow, tearing up, (wet clothes: 1.5), starry sky print, long tops, long sleeves, witch hat, cape, (pouting: 1.5), (floating: 1.3), sun light, reflection, (sweating: 1.55), sparkle, on back, top down view, full body
Negative: lowres, bad anatomy, bad hands, text, error, missing fingers, extra digit, fewer digits, cropped, worst quality, low quality, normal quality, jpeg artifacts, signature, watermark, username, blurry, artist name
On the note of the character, I would suggest her tag being named amber_(rommilia) instead, since 1). OCs are usually tagged with the owner/creator's name as the qualifier and 2). So other users can immediately identify her to be your character.
On the note of the character, I would suggest her tag being named amber_(rommilia) instead, since 1). OCs are usually tagged with the owner/creator's name as the qualifier and 2). So other users can immediately identify her to be your character.
Either my shit is bootleg or I'm doing something wrong. I can't seem to replicate 95% of this work. Or should I even be able to? What about the rest of you?
(prompt might be off a bit because my history tab doesnt show up anymore)
prompt: masterpiece, best quality. a Camilla-FE.bin in a dress sitting at a table, by artgerm. ((hair over one eye)), purple hair, (large breasts), long hair, (thick thighs), solo, cleavage, detailed face, character focus, absurd res, detailed eyes, sexy face, [intricate details], [[photorealistic]], detailed facial features, natural lighting, sharp focus, [makeup], [[high heels]] , looking at viewer, [restaurant], indoors, [[nighttime]], purple eyes, (feet out of frame), strong legs, muscular legs
I love the way it matched the color scheme on this. Sometimes when I use NovelAI, I feel like I'm 'playing dress up' with the character but it's just so much fun.
Firstpass 0x0 just means it's automatic and that's what I'm using too. I have one weird idea left. When mixing, set model A as trinart2_step115000, B as nai-animefull-final-pruned and multiplier to 0.8. You can also check "save as float16" even though I honestly doubt it changes much.
It's "better anatomy" but yes that's the one. I mixed it 50% with r34_e4 and then 0.65 of the result with trinart. Finally I did nai * 0.7 + the_previous_mix * 0.3. Not sure if it changes much though, I was just messing around. Results are very similar to a nai * 0.8 + trinart * 0.2.
Thank you very much. I hope the drawing of ai will be more abundant
Is the f111 model used in this picture the same as 'Zeifer F111 Female Nude (better animation)? Thank you.
It's "better anatomy" but yes that's the one. I mixed it 50% with r34_e4 and then 0.65 of the result with trinart. Finally I did nai * 0.7 + the_previous_mix * 0.3. Not sure if it changes much though, I was just messing around. Results are very similar to a nai * 0.8 + trinart * 0.2.
realistic, masterpiece, high quality, highres, cute afraid loli goblin goat slave with wavy hair, cleavage cutout, small breasts, detailed face and eyes, (frills:1.2) (solo focus:1.2) pov, aroused, embarrassed, antlers, by Jeremy Lipking, by Antonio J. Manzanedo, by (Alphonse Mucha) digital painting Negative prompt: lowres, bad anatomy, bad hands, text, error, missing fingers, extra digit, fewer digits, cropped, worst quality, low quality, normal quality, jpeg artifacts, signature, watermark, username, blurry, out of focus, censorship, ugly, old, deformed, amateur drawing, odd, fat, adult
Steps: 20, Sampler: DDIM, CFG scale: 14, Seed: 1404394528, Size: 512x960, Model hash: 45a4bfe3, Denoising strength: 0.6, First pass size: 0x0 highres fix Upscale latent space image when doing hires. fix Stop At last layers of CLIP model:1
Still got a different image - -! Maybe "First pass size: 0x0" cause?
Firstpass 0x0 just means it's automatic and that's what I'm using too. I have one weird idea left. When mixing, set model A as trinart2_step115000, B as nai-animefull-final-pruned and multiplier to 0.8. You can also check "save as float16" even though I honestly doubt it changes much.
Must be a change in how exactly the models are merged but it shouldn't matter anyway. I did a merge on the current version of the ui and got the same hash as yours [45a4bfe3]. It still generates the same image. I used highres fix with latent space upscale and CLIP:1 so maybe it's one of these settings?
My parameters:
realistic, masterpiece, high quality, highres, cute afraid loli goblin goat slave with wavy hair, cleavage cutout, small breasts, detailed face and eyes, (frills:1.2) (solo focus:1.2) pov, aroused, embarrassed, antlers, by Jeremy Lipking, by Antonio J. Manzanedo, by (Alphonse Mucha) digital painting Negative prompt: lowres, bad anatomy, bad hands, text, error, missing fingers, extra digit, fewer digits, cropped, worst quality, low quality, normal quality, jpeg artifacts, signature, watermark, username, blurry, out of focus, censorship, ugly, old, deformed, amateur drawing, odd, fat, adult
Steps: 20, Sampler: DDIM, CFG scale: 14, Seed: 1404394528, Size: 512x960, Model hash: 45a4bfe3, Denoising strength: 0.6, First pass size: 0x0 highres fix Upscale latent space image when doing hires. fix Stop At last layers of CLIP model:1
Still got a different image - -! Maybe "First pass size: 0x0" cause?
